MS Regressions
Erica Baller
11/18/2021
Dac summary
Welcome to the MS project! This is a really exciting project dedicated to understanding psychopathology, particular depression in patients with MS. It is unqiue in a variety of ways, primarily that it uses clinical images that are research grade.
The team so far is big. PIs- Ted Satterthwaite and Taki Shinohara Senior Scientists - Azeez Adebimpe and Matt Cieslak Research Specialist- Timothy Robert-Fitzgerald Graduate students - Melissa Martin Data Analysist - Sidney Covitz Collaborators - Aaron Alexander-Bloch and Jenna Young DAC Pull done by - Victoria Rautman(2020/2021) and Sunil Thomas (2018)
Date of FINAL Pull: 1/19/2021 Spreadsheet: /Users/eballer/BBL/msdepression/data/dac/investigatingdepressioninmspatients_dates_right_format.csv Summary of request to Victoria @ DAC: — 9/2/2020
DAC Request Non-funded IRB Approved Research (enter IRB number): 843669 Criteria Display? Description / Exclusions / Limitations / Filters MRN Y Unique identifier requested, but can be masked MRN Visit ID Y
Patient Class(es) Please select only which class(es) you will need. x Inpatient X Outpatient x Emergency Age or DOB ranges Y
Gender Y
Race Y
Department(s) Provide department numbers not just names. N
Provider(s) Provide ID’s not just names. N
Date(s) Include in the specific range and date types (eg, admit, order, result) 1/1/2010- 12/31/20 All scans based on date of first treatment with glatiramer acetate Procedure Please include the specific procedure codes. (ICD9/ICD10 is preferred for inpatient) Y 88.91 (Magnetic Resonance Imaging of Brain)
Diagnosis Please include the specific ICD-9/ICD-10 codes including all decimal points. Do not simply include ranges or wildcards. Y Multiple sclerosis (ICD-9: 340; ICD-10: G35), Depressive disorders (ICD-9: 296.99, 296.21, 296.22, 296.23, 296.24, 296.25, 296.26, 296.20, 296.31, 296.32, 296.33, 296.34, 296.35, 296.26, 296.30, 300.4, 293.83, 311, ICD-10: F32.0, F32.1, F32.2, F32.3, F32.4, F32.5, F32.6, F33.0, F33.1, F33.2, F33.41, F33.42, F33.9, F34.1, F06.31, F06.32, F06.34, F32.8, F32.9, major depressive disorder, persistent depressive disorder, depressive disorder due to another medical condition, other specified depressive disorder, unspecified depressive disorder) Orders N
Medication Please list as it is ordered within the UPHS EMR’s – medication id’s preferred Y Glatiramer Acetate (Copaxone) Lab Result Please list the lab as it is ordered within the UPHS EMR’s. Y
Other Y EDSS scores (if present) Other Y All medications patients are receiving Other Y All PACS accession numbers available Other Y PHQ-2 and PHQ-9 if present — 9/16/2020: Per her note to me: We also specified that we wanted to pull images with ORIG_CPT values of 70551, 70552, 70553, MHDI, IMGMR0128, and we wanted scans with department ID 361, 547, 6004, or 6013. And I may have mentioned this before but the ICD9 and ICD10 codes for MS patients are 340 and G35, respectively
*For us, we are additionally interested in ALL ICD9/10 codes, ALL medications they are on at the time of scan, specific labs (as indicated), EDSS , PHQ-9 and 2 scores. The particular reason for this is that we are going to be looking at the relationship of brain imaging to depression in the MS group.
Mock columns: DE_PAT_ID PAT_ID EMPI HUP_MRN PAT_NM_WMRN SEX RACE ETHNIC_GROUP ORDER_ID ORDERING_DTTM BEGIN_EXAM_DTTM END_EXAM_DTTM TECH_USER_ID TECHNOLOGIST ACCESSION_NUM PROC_ID PROC_NAME PROC_CODE ORIG_CPT PERFORMING_CSN_ID COVERAGE_ID FIN_CLASS_NAME VISIT_EPM_ID PERFORMING_DEP_ID DEPARTMENT_NAME MODALITY ASSOCIATED_ICD9 ASSOCIATED_ICD10 PAT_AGE_AT_EXAM MRI_ENC_AGE CURRENT_AGE All ICD-9 or 10 codes All medication codes wbc rbc hemoglobin csf studies… phq-9 score (and date) phq-2 score (and date) b12 folate TSH RPR Vit D
| 1/8/2021 email update (me to Victoria Rautman): What we realized is that the best way to define inclusion for the MS group is to focus on people who were seen at some point in neurology clinic. |
| If you’d be so kind, I’d be grateful for *hopefully the last pull. We would use the previous one, but instead of using the ICD 9/10 codes associated with the actual scan, we would use the location of clinic visits, and pull ALL scans, before or after diagnosis. |
| New updates: -Pull all the scans for patients who have ever been seen by “NEUROLOGY SOUTH PAVILLION” and “NEUROLOGY HUP” departments for clinic. |
| -Please include provider name associated with these department visits. |
| -Don’t worry about doing exclusions based on ICD9/10 codes for the scans themselves, please include all and we will sort through it ourselves |
| Please also include vitamin D levels in addition to other labs |
| The rest of the query would be the same as the most recent query you sent back. |
Radiology sent images to Taki and group 4/2021
This script goes through data that has been returned from the DAC on 1/19 and summarizes it
homedir <- "/Users/eballer/BBL/msdepression/"
##########################################################
#### Prepare the date ####
##########################################################
#load our df
#from Jan 2021 pull
data <- read.csv(paste0(homedir,"/data/dac/investigatingdepressioninmspatients_dates_right_format.csv"), header = TRUE, stringsAsFactors = FALSE, na.strings = c("NULL"), sep =",")
ms_providers <- c("MARKOWITZ, CLYDE E.", "JACOBS, DINA A.", "WILLIAMSON, ERIC MICHAEL-LEE", "BERGER, JOSEPH ROBERT", "BERGER DO, JOSEPH", "PRUITT, AMY A.", "KOLSON, DENNIS L.", "CHAHIN, SALIM", "NARULA, SONA", "BAR-OR, AMIT")
columns_to_make_integer <- c("ACCESSION_NUM", "PAT_AGE_AT_EXAM","MRI_ENC_AGE","CURRENT_AGE","hemoglobin", "WBC","RBC","B12","FOLATE", "TSH", "RPR","CSFAPPEAR1","CSFAPPEAR4","CSFCOLOR1", "CSFCOLOR4","CSFTUBE","CSFTUBE4","VitD","PHQ.2","PHQ.9")
##########################
### some preprocessing ###
##########################
#we need to keep scans from people seen by an MS provider (n = 17067) who have an MS diagnostic code (n=16,830)
#we make a bunch of columns from character to integer, binarize sex and race, and put date into a nice format YYYYMMDD, the %m%d%y indicates what it started out as
#goal is to keep people who were seen by
data_empi_acc_f_phq <- data %>%
filter(Provider %in% ms_providers) %>% #n=17,067
filter(grepl("G35", ICD10)) %>% #n=16830
mutate(across(.cols = columns_to_make_integer, .fns = as.integer)) %>%
mutate(sex_binarized = ifelse(SEX == "MALE", 1, 2)) %>%
mutate(osex = ordered(sex_binarized,levels = c(1,2), labels = c("Male","Female"))) %>%
mutate(race_binarized = ifelse(RACE == "WHITE", 1, 2)) %>%
mutate(orace = ordered(race_binarized,levels = c(1,2), labels = c("White","Non-white"))) %>%
mutate(EXAM_DATE = as.Date(BEGIN_EXAM_DTTM, format = "%m/%d/%y")) %>%
mutate(EXAM_DATE = gsub(EXAM_DATE, pattern = "-", replacement = ""))## Note: Using an external vector in selections is ambiguous.
## ℹ Use `all_of(columns_to_make_integer)` instead of `columns_to_make_integer` to silence this message.
## ℹ See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
## This message is displayed once per session.## Warning in mask$eval_all_mutate(quo): NAs introduced by coercion
## Warning in mask$eval_all_mutate(quo): NAs introduced by coercion
## Warning in mask$eval_all_mutate(quo): NAs introduced by coercion
## Warning in mask$eval_all_mutate(quo): NAs introduced by coercion
## Warning in mask$eval_all_mutate(quo): NAs introduced by coercion
## Warning in mask$eval_all_mutate(quo): NAs introduced by coercion
## Warning in mask$eval_all_mutate(quo): NAs introduced by coercion
## Warning in mask$eval_all_mutate(quo): NAs introduced by coercion
## Warning in mask$eval_all_mutate(quo): NAs introduced by coercion##########################################################
#### Step 1- make F3* depressed and healthy groups* ######
##########################################################
### This is identical to up above, but repeated so I can run this chunk separately
########
### separate out depressed group
########
print("separate out depressed group")## [1] "separate out depressed group"dep_df <- data_empi_acc_f_phq %>% filter(grepl("F3", ICD10)) #n = 2123
num_depression <- dim(dep_df)[1]
num_unique_EMPI_and_dep <- length(unique(dep_df$EMPI)) #n = 493
print(paste0("N with ICD 9/10 codes for depression = ", num_depression))## [1] "N with ICD 9/10 codes for depression = 2123"print(paste0("N with ICD 9/10 codes for depression and unique EMPI in whole depression group = ", num_unique_EMPI_and_dep, " out of ", num_depression))## [1] "N with ICD 9/10 codes for depression and unique EMPI in whole depression group = 493 out of 2123"#find number of participants who meet criteria for depression in the PHQ2/9 symptom rating
print(paste0("N with ICD 9/10 codes for depression, PHQ2 >= 3, unique = ", length(!is.na(unique(dep_df$EMPI[dep_df$PHQ.2 >= 3]))))) #n=23## [1] "N with ICD 9/10 codes for depression, PHQ2 >= 3, unique = 23"print(paste0("N with ICD 9/10 codes for depression, PHQ9 >= 10, unique = ", length(!is.na(unique(dep_df$EMPI[dep_df$PHQ.9 >= 10])))))#n=26## [1] "N with ICD 9/10 codes for depression, PHQ9 >= 10, unique = 26"#######
### separate out healthy group
#######
print("separate out healthy group")## [1] "separate out healthy group"healthy_df <- data_empi_acc_f_phq %>% filter(!grepl("F3", ICD10)) #n = 14707
num_healthy <- dim(healthy_df)[1]
num_unique_EMPI_and_healthy <- length(unique(healthy_df$EMPI)) #n = 3244
print(paste0("N with WITHOUT depression ICD 9/10 codes (i.e. \"healthy\"): ", num_healthy))## [1] "N with WITHOUT depression ICD 9/10 codes (i.e. \"healthy\"): 14707"print(paste0("N with WITHOUT ICD 9/10 depression codes for depression and unique EMPI = ", num_unique_EMPI_and_healthy, " out of ", num_healthy))## [1] "N with WITHOUT ICD 9/10 depression codes for depression and unique EMPI = 3244 out of 14707"##########################################################
#### Step 2- Identify PHQ2 >=3 in healthy group ######
##########################################################
print("Get people in healthy group with phq2s for dep/healthy")## [1] "Get people in healthy group with phq2s for dep/healthy"# num in healthy group with PHQ2
phq2_subset <- healthy_df %>% filter(!is.na(PHQ.2)) #n=3646, unique 698
depressed_phq2_subset <- phq2_subset %>% filter(PHQ.2 >= 3) #n = 47
depressed_unique_phq2_empis <- unique(depressed_phq2_subset$EMPI) #n=10
healthy_phq2_subset <- phq2_subset %>% filter(PHQ.2 == 0) #n = 3326
healthy_unique_phq2_empis <- unique(healthy_phq2_subset$EMPI) #n = 631
print(paste0("PHQ2 breakdown in healthy group (total n with score = ", dim(phq2_subset)[1], "[unique = ", length(unique(phq2_subset$EMPI)),"]): PHQ2 >= 3 :", dim(depressed_phq2_subset)[1], "[unique = ", length(depressed_unique_phq2_empis), "]; PHQ2 == 0 : ", dim(healthy_phq2_subset)[1], "[unique = ", length(healthy_unique_phq2_empis),"]")) #PHQ2 breakdown in healthy group (total n with score = 3646[unique = 698]): PHQ2 >= 3 :47[unique = 10]; PHQ2 == 0 : 3326[unique = 631]## [1] "PHQ2 breakdown in healthy group (total n with score = 3646[unique = 698]): PHQ2 >= 3 :47[unique = 10]; PHQ2 == 0 : 3326[unique = 631]"##########################################################
#### Step 3- Identify PHQ9 >=10 in healthy group ######
##########################################################
print("Get people in healthy group with phq9s for dep/healthy")## [1] "Get people in healthy group with phq9s for dep/healthy"# num in healthy group with PHQ9
phq9_subset <- healthy_df %>% filter(!is.na(PHQ.9)) #n=121 (unique 5)
depressed_phq9_subset <- phq9_subset %>% filter(PHQ.9 >= 10) #n = 32
depressed_unique_phq9_empis <- unique(depressed_phq9_subset$EMPI) #n=9
healthy_phq9_subset <- phq9_subset %>% filter(PHQ.9 == 0) #n = 21
healthy_unique_phq9_empis <- unique(healthy_phq9_subset$EMPI) #n = 4
print(paste0("PHQ9 breakdown in healthy group (total n with score = ", dim(phq9_subset)[1], "[unique = ", length(unique(phq9_subset$EMPI)),"]): PHQ9 >= 10 :", dim(depressed_phq9_subset)[1], "[unique = ", length(depressed_unique_phq9_empis), "]; PHQ9 == 0 : ", dim(healthy_phq9_subset)[1], "[unique = ", length(unique(healthy_phq9_subset$EMPI)),"]"))## [1] "PHQ9 breakdown in healthy group (total n with score = 121[unique = 25]): PHQ9 >= 10 :32[unique = 9]; PHQ9 == 0 : 21[unique = 4]"##########################################################
#Step 4 - Combine for depressed group and healthy groups #
##########################################################
print("Combine groups for new cohort")## [1] "Combine groups for new cohort"gain_in_dep_group <- length(depressed_unique_phq2_empis) + length(depressed_unique_phq9_empis)
dep_num_icd10_plus_phq2_and_phq9 <- num_unique_EMPI_and_dep + gain_in_dep_group # we can do this because we know they are each exclusive - num_unique_empi_and_dep from icd10 group, and no overlap in phq2 and 9 group
print(paste0("number of UNIQUE people with an icd10 code for Depression OR scored positive on PHQ2(>=3) or PHQ9(>=10) : ", dep_num_icd10_plus_phq2_and_phq9, ", for a gain of n = ", gain_in_dep_group))## [1] "number of UNIQUE people with an icd10 code for Depression OR scored positive on PHQ2(>=3) or PHQ9(>=10) : 512, for a gain of n = 19"healthy_num_no_icd10_plus_phq2_and_phq9 <- num_unique_EMPI_and_healthy - gain_in_dep_group #take number of people who don't have depression diagnosis and don't have a depression phq2/9
print(paste0("number of people with NO icd10 code for Depression and NEVER had a PHQ2(>=3) or PHQ9(>=10) : ", healthy_num_no_icd10_plus_phq2_and_phq9, ", for a loss of n = ", gain_in_dep_group))## [1] "number of people with NO icd10 code for Depression and NEVER had a PHQ2(>=3) or PHQ9(>=10) : 3225, for a loss of n = 19"healthy_num_no_icd10_plus_phq2_and_phq9_MUST_HAVE_PHQ2_OR_9 <- length(healthy_unique_phq2_empis) + length(healthy_unique_phq9_empis)
print(paste0("number of UNIQUE people with NO icd10 code for Depression and HAS HAD AT LEAST 1 PHQ2 or 9 that was 0) : ", healthy_num_no_icd10_plus_phq2_and_phq9_MUST_HAVE_PHQ2_OR_9))## [1] "number of UNIQUE people with NO icd10 code for Depression and HAS HAD AT LEAST 1 PHQ2 or 9 that was 0) : 635"print(paste0("Cleanest cohort (depressed in icd9/10 or Phq2/9; healthy by at least one phq2/9 and never depression dx UNIQUE ---->>>> Depressed: ", dep_num_icd10_plus_phq2_and_phq9, "; Healthy: ", healthy_num_no_icd10_plus_phq2_and_phq9_MUST_HAVE_PHQ2_OR_9))## [1] "Cleanest cohort (depressed in icd9/10 or Phq2/9; healthy by at least one phq2/9 and never depression dx UNIQUE ---->>>> Depressed: 512; Healthy: 635"##### THESE ARE THE DATAFRAMES I WANT TO USE GOING FORWARD ###
#all empis for depressed group include people with depresssion dx, and people who got in for phq2 or phq9
empis_for_depressed_group <- append(
append(unique(dep_df$EMPI), depressed_unique_phq2_empis), depressed_unique_phq9_empis)
final_depressed_group_withICD_AND_depressed_phq2_or_9 <-
data_empi_acc_f_phq %>%
filter(EMPI %in% empis_for_depressed_group)
#------
empis_for_healthy_group <- append(healthy_unique_phq2_empis, healthy_unique_phq9_empis)
final_healthy_group_withNOICD_AND_phq2_or_9_0<-
data_empi_acc_f_phq %>%
filter(EMPI %in% empis_for_healthy_group)
#------
### add a column to the original dataframe with a 0 if not in any group, -1 if healthy, and 1 if depressed
data_empi_acc_f_phq$depGroupVar = 0
data_empi_acc_f_phq$depGroupVar[which(data_empi_acc_f_phq$EMPI %in% empis_for_depressed_group)] = 1
data_empi_acc_f_phq$depGroupVar[which(data_empi_acc_f_phq$EMPI %in% empis_for_healthy_group)] = -1
data_empi_acc_f_phq$depGroupVar = as.factor(data_empi_acc_f_phq$depGroupVar)
#------ Group for Tim to check quality issues ----------
write.csv(data_empi_acc_f_phq, "/Users/eballer/BBL/msdepression/data/dac/data_empi_acc_f_q.csv", col.names = T, sep = ",")## Warning in write.csv(data_empi_acc_f_phq, "/Users/eballer/BBL/msdepression/data/
## dac/data_empi_acc_f_q.csv", : attempt to set 'col.names' ignored## Warning in write.csv(data_empi_acc_f_phq, "/Users/eballer/BBL/msdepression/data/
## dac/data_empi_acc_f_q.csv", : attempt to set 'sep' ignoredsubset <- subset(data_empi_acc_f_phq, select = c("ACCESSION_NUM","EMPI","EXAM_DATE","depGroupVar"))
melissa_qc_data <- read.csv("/Users/eballer/BBL/msdepression/data/dac/just_empi_date_rating", sep = ",", header = F) #n = 2841
names(melissa_qc_data) <- c("EMPI", "EXAM_DATE", "rating")
bad_flairs <- read.csv("/Users/eballer/BBL/msdepression/data/dac/missing_flair.csv", sep = ",", header = T) #n = 1561
bad_flairs$EXAM_DATE <- gsub(pattern = "-", replacement = "", x = bad_flairs$date)
subset_with_qc <- merge(subset, melissa_qc_data, by = c("EMPI", "EXAM_DATE"))
subset_excluded <- merge(subset, bad_flairs, by = c("EMPI", "EXAM_DATE")) #n = 1973; healthy = 457, uncategorized = 1284, depressed = 232
unique <- subset_excluded %>%
group_by(EMPI) %>%
arrange(EXAM_DATE) %>%
slice(1) %>%
ungroup() #n (total) = 458, healthy = 114, uncat = 282. dep = 62
write.csv(subset_with_qc, "/Users/eballer/BBL/msdepression/data/dac/subset_with_qc_for_tim_20211117.csv", col.names = T)## Warning in write.csv(subset_with_qc, "/Users/eballer/BBL/msdepression/data/dac/
## subset_with_qc_for_tim_20211117.csv", : attempt to set 'col.names' ignoredwrite.csv(subset_with_qc, "/Users/eballer/BBL/msdepression/data/dac/subset_bad_flairs_for_tim_20211117.csv", col.names = T)## Warning in write.csv(subset_with_qc, "/Users/eballer/BBL/msdepression/data/dac/
## subset_bad_flairs_for_tim_20211117.csv", : attempt to set 'col.names' ignored##########################################################
# Step 5 - Read in Fascicle info #
##########################################################
fascicle_proportions <- read.csv(paste0(homedir, "/results/fascicle_volumes_all_subjects_roi_n2336.csv"), header = T, sep = ",") #n = 2336
#fascicle names are contained in all but the first 2 columns of the fascicle_proportions df
fascicle_names <- names(fascicle_proportions[3:dim(fascicle_proportions)[2]])
write.table(fascicle_names,"/Users/eballer/BBL/msdepression/templates/dti/HCP_YA1065_tractography/fascicle_names.csv", row.names = F, quote = F, col.names = F)
#fascicle_mapping, returns numerical mapping as well as names of tracts
fascicle_bundle_mapping <- get_fascicle_bundle_mapping()
##########################################################
# Step 6 - Merge with data_empi #
##########################################################
df_demo_and_fascicles <- merge(data_empi_acc_f_phq, fascicle_proportions, by = c("EMPI", "EXAM_DATE")) #n = 2962
##########################################################
# Step 7 - Histos #
##########################################################
##### Histograms for PHQ2/9 healthy and depressed
hist(df_demo_and_fascicles$PHQ.2[which(!is.na(df_demo_and_fascicles$PHQ.2))], main = paste0("PHQ-2 In Good Mimosa Group : n = ", length(which(!is.na(df_demo_and_fascicles$PHQ.2)))), xlab = "PHQ-2", breaks = 8)
hist(df_demo_and_fascicles$PHQ.9[which(!is.na(df_demo_and_fascicles$PHQ.9))], main = paste0("PHQ-9 In Good Mimosa Group : n = ", length(which(!is.na(df_demo_and_fascicles$PHQ.9)))), xlab = "PHQ-9", breaks = 8) 
hist(df_demo_and_fascicles$PHQ.2[which(!is.na(df_demo_and_fascicles$PHQ.2) & (df_demo_and_fascicles$depGroupVar == 1))], main = paste0("PHQ-2/Depressed In Good Mimosa Group : n = ", length(which(!is.na(df_demo_and_fascicles$PHQ.2) & df_demo_and_fascicles$depGroupVar == 1))), xlab = "PHQ-2", breaks = 8)
hist(df_demo_and_fascicles$PHQ.9[which(!is.na(df_demo_and_fascicles$PHQ.9) & (df_demo_and_fascicles$depGroupVar == 1))], main = paste0("PHQ-9/Depressed In Good Mimosa Group : n = ", length(which(!is.na(df_demo_and_fascicles$PHQ.9) & df_demo_and_fascicles$depGroupVar == 1))), xlab = "PHQ-9", breaks = 8)
hist(df_demo_and_fascicles$PHQ.2[which(!is.na(df_demo_and_fascicles$PHQ.2) & (df_demo_and_fascicles$depGroupVar == -1))], main = paste0("PHQ-2/Healthy In Good Mimosa Group : n = ", length(which(!is.na(df_demo_and_fascicles$PHQ.2) & df_demo_and_fascicles$depGroupVar == -1))), xlab = "PHQ-2", breaks = 8)
hist(df_demo_and_fascicles$PHQ.9[which(!is.na(df_demo_and_fascicles$PHQ.9) & (df_demo_and_fascicles$depGroupVar == -1))], main = paste0("PHQ-9/Healthy In Good Mimosa Group : n = ", length(which(!is.na(df_demo_and_fascicles$PHQ.9) & df_demo_and_fascicles$depGroupVar == -1))), xlab = "PHQ-9", breaks = 8)
Step 8 - Visualize all Tracts
##########################################################
# Step 8 - Visualize all Tracts #
##########################################################
just_dep_and_empi <- subset(data_empi_acc_f_phq, select = c("EMPI", "depGroupVar"))
added_dep_to_fascicles <- merge(just_dep_and_empi, fascicle_proportions, by = c("EMPI"))
melted_df <- melt(added_dep_to_fascicles, id.vars = c("EMPI", "EXAM_DATE", "depGroupVar"))
lesioned <- subset(melted_df, value > 0)
lesioned_dx <- subset(lesioned, depGroupVar != 0)
healthy <- subset(lesioned, depGroupVar == -1)
depressed <- subset(lesioned, depGroupVar == 1)
q<-ggplot(lesioned, aes(x=value, fill=depGroupVar)) + geom_histogram() + facet_wrap(~variable)
r<-ggplot(lesioned_dx, aes(x=value, fill=depGroupVar)) + geom_histogram() + facet_wrap(~variable)
x<-ggplot(lesioned, aes(x=value)) + geom_histogram() + facet_wrap(~variable)
y<-ggplot(healthy, aes(x=value)) + geom_histogram() + facet_wrap(~variable)
z<-ggplot(depressed, aes(x=value)) + geom_histogram() + facet_wrap(~variable)
print(q)## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
print(r)## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
print(x)## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
print(y)## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
print(z)## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Step 9 - Regressions
##########################################################
# Step 9 - Regressions #
##########################################################
#isolate out only the depressed group and healthy group
dep_and_healthy_groups_for_ICD_analysis <- df_demo_and_fascicles[df_demo_and_fascicles$depGroupVar != 0,] #n depressed = 393, healthy = 713, total n = 1106
#lm
fascicle_lm <- lapply(fascicle_names, function(x)
{
lm(substitute(i ~ depGroupVar, list(i = as.name(x))), data = dep_and_healthy_groups_for_ICD_analysis)
})
names(fascicle_lm) <- fascicle_names
#anova
fascicle_anova <- lapply(fascicle_lm, anova)
#fdr corrected
fascicle_anova_fdr <- fdr_anova_generic(fascicle_anova, 1)## p_anova
## AF_L 2.578214e-03
## AF_R 6.174158e-03
## C_FPH_L 9.624075e-01
## C_FPH_R 5.844313e-01
## C_FP_L 3.398230e-02
## C_FP_R 5.158020e-01
## C_PH_L 1.357335e-03
## C_PHP_L 9.659307e-02
## C_PHP_R 2.603931e-02
## C_PH_R 4.552169e-01
## C_R_L 1.945615e-02
## C_R_R 1.028564e-01
## EMC_L 3.373818e-02
## EMC_R 9.679307e-02
## FAT_L 2.282320e-02
## FAT_R 2.562681e-05
## IFOF_L 1.121774e-03
## IFOF_R 4.455661e-02
## ILF_L 4.880404e-05
## ILF_R 8.276630e-03
## MdLF_L 3.453679e-02
## MdLF_R 3.176372e-02
## PAT_L 3.443343e-03
## PAT_R 3.748231e-05
## SLF1_L 6.381944e-04
## SLF1_R 3.854758e-01
## SLF2_L 2.085115e-02
## SLF2_R 6.031135e-04
## SLF3_L 5.061243e-03
## SLF3_R 3.607070e-03
## UF_L 1.099838e-01
## UF_R 1.214621e-01
## VOF_L 3.322972e-02
## VOF_R 5.213149e-06
## CB_L 9.309656e-01
## CB_R 5.519092e-01
## ICP_L 5.679475e-01
## ICP_R 8.055311e-01
## MCP 7.730769e-01
## SCP 1.778219e-01
## V 6.658906e-01
## CNIII_L 6.343637e-01
## CNIII_R 4.666709e-01
## CNII_L 3.700836e-01
## CNII_R 3.529199e-01
## CNVIII_L 2.361714e-01
## CNVIII_R 7.577010e-01
## CNVII_L 2.101867e-01
## CNVII_R 2.937322e-01
## CNV_L 6.910983e-02
## CNV_R 3.083527e-03
## AR_L 4.026356e-03
## AR_R 9.457673e-05
## CBT_L 3.048118e-05
## CBT_R 5.687075e-03
## CPT_F_L 1.886671e-03
## CPT_F_R 5.768941e-05
## CPT_O_L 4.502135e-04
## CPT_O_R 4.413450e-02
## CPT_P_L 4.252117e-02
## CPT_P_R 3.083369e-04
## CS_A_L 1.065540e-03
## CS_A_R 6.444177e-03
## CS_P_L 2.023262e-02
## CS_P_R 1.146695e-01
## CS_S_L 6.937632e-04
## CS_S_R 8.898003e-05
## CST_L 2.073714e-02
## CST_R 8.125128e-04
## DRTT_L 4.802059e-03
## DRTT_R 4.888754e-06
## F_L 3.842894e-03
## F_R 1.112795e-01
## ML_L 1.425907e-01
## ML_R 6.727661e-05
## OR_L 9.572355e-03
## OR_R 8.179265e-03
## RST_L 3.462419e-03
## RST_R 3.550929e-05
## TR_A_L 2.192602e-04
## TR_A_R 1.347576e-05
## TR_P_L 1.056143e-03
## TR_P_R 3.014904e-04
## TR_S_L 1.229346e-03
## TR_S_R 3.440764e-07
## AC 4.324205e-02
## CC 6.941672e-03print(fascicle_anova_fdr)## component p_FDR_corr
## 1 AF_L 0.008
## 2 AF_R 0.014
## 3 C_PH_L 0.005
## 4 C_PHP_R 0.046
## 5 C_R_L 0.038
## 6 FAT_L 0.041
## 7 FAT_R 0
## 8 IFOF_L 0.004
## 9 ILF_L 0
## 10 ILF_R 0.017
## 11 PAT_L 0.01
## 12 PAT_R 0
## 13 SLF1_L 0.003
## 14 SLF2_L 0.039
## 15 SLF2_R 0.003
## 16 SLF3_L 0.012
## 17 SLF3_R 0.01
## 18 VOF_R 0
## 19 CNV_R 0.009
## 20 AR_L 0.01
## 21 AR_R 0.001
## 22 CBT_L 0
## 23 CBT_R 0.013
## 24 CPT_F_L 0.006
## 25 CPT_F_R 0.001
## 26 CPT_O_L 0.002
## 27 CPT_P_R 0.002
## 28 CS_A_L 0.004
## 29 CS_A_R 0.014
## 30 CS_P_L 0.039
## 31 CS_S_L 0.003
## 32 CS_S_R 0.001
## 33 CST_L 0.039
## 34 CST_R 0.003
## 35 DRTT_L 0.012
## 36 DRTT_R 0
## 37 F_L 0.01
## 38 ML_R 0.001
## 39 OR_L 0.019
## 40 OR_R 0.017
## 41 RST_L 0.01
## 42 RST_R 0
## 43 TR_A_L 0.001
## 44 TR_A_R 0
## 45 TR_P_L 0.004
## 46 TR_P_R 0.002
## 47 TR_S_L 0.004
## 48 TR_S_R 0
## 49 CC 0.015#fdr corrected only association cortex
fascicle_anova_w_fiber_mapping <- fascicle_anova[which(fascicle_bundle_mapping$name_vector == "association")]
fascicle_anova_fdr_association <- fdr_anova_generic(fascicle_anova_w_fiber_mapping,1)## p_anova
## AF_L 2.578214e-03
## AF_R 6.174158e-03
## C_FPH_L 9.624075e-01
## C_FPH_R 5.844313e-01
## C_FP_L 3.398230e-02
## C_FP_R 5.158020e-01
## C_PH_L 1.357335e-03
## C_PHP_L 9.659307e-02
## C_PHP_R 2.603931e-02
## C_PH_R 4.552169e-01
## C_R_L 1.945615e-02
## C_R_R 1.028564e-01
## EMC_L 3.373818e-02
## EMC_R 9.679307e-02
## FAT_L 2.282320e-02
## FAT_R 2.562681e-05
## IFOF_L 1.121774e-03
## IFOF_R 4.455661e-02
## ILF_L 4.880404e-05
## ILF_R 8.276630e-03
## MdLF_L 3.453679e-02
## MdLF_R 3.176372e-02
## PAT_L 3.443343e-03
## PAT_R 3.748231e-05
## SLF1_L 6.381944e-04
## SLF1_R 3.854758e-01
## SLF2_L 2.085115e-02
## SLF2_R 6.031135e-04
## SLF3_L 5.061243e-03
## SLF3_R 3.607070e-03
## UF_L 1.099838e-01
## UF_R 1.214621e-01
## VOF_L 3.322972e-02
## VOF_R 5.213149e-06print(fascicle_anova_fdr_association)## component p_FDR_corr
## 1 AF_L 0.01
## 2 AF_R 0.016
## 3 C_PH_L 0.006
## 4 C_PHP_R 0.049
## 5 C_R_L 0.044
## 6 FAT_L 0.046
## 7 FAT_R 0
## 8 IFOF_L 0.005
## 9 ILF_L 0
## 10 ILF_R 0.02
## 11 PAT_L 0.011
## 12 PAT_R 0
## 13 SLF1_L 0.004
## 14 SLF2_L 0.044
## 15 SLF2_R 0.004
## 16 SLF3_L 0.014
## 17 SLF3_R 0.011
## 18 VOF_R 0#visreg
sapply(fascicle_lm, visreg)
## AF_L AF_R C_FPH_L C_FPH_R C_FP_L
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## C_FP_R C_PH_L C_PHP_L C_PHP_R C_PH_R
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## C_R_L C_R_R EMC_L EMC_R FAT_L
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## FAT_R IFOF_L IFOF_R ILF_L ILF_R
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## MdLF_L MdLF_R PAT_L PAT_R SLF1_L
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## SLF1_R SLF2_L SLF2_R SLF3_L SLF3_R
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## UF_L UF_R VOF_L VOF_R CB_L
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## CB_R ICP_L ICP_R MCP SCP
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## V CNIII_L CNIII_R CNII_L CNII_R
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## CNVIII_L CNVIII_R CNVII_L CNVII_R CNV_L
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## CNV_R AR_L AR_R CBT_L CBT_R
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## CPT_F_L CPT_F_R CPT_O_L CPT_O_R CPT_P_L
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## CPT_P_R CS_A_L CS_A_R CS_P_L CS_P_R
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## CS_S_L CS_S_R CST_L CST_R DRTT_L
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## DRTT_R F_L F_R ML_L ML_R
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## OR_L OR_R RST_L RST_R TR_A_L
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## TR_A_R TR_P_L TR_P_R TR_S_L TR_S_R
## fit data.frame,5 data.frame,5 data.frame,5 data.frame,5 data.frame,5
## res data.frame,4 data.frame,4 data.frame,4 data.frame,4 data.frame,4
## meta list,6 list,6 list,6 list,6 list,6
## AC CC
## fit data.frame,5 data.frame,5
## res data.frame,4 data.frame,4
## meta list,6 list,6#isolate out only the depressed group and healthy group
dep_and_healthy_groups_for_ICD_analysis <- df_demo_and_fascicles[df_demo_and_fascicles$depGroupVar != 0,] #n depressed = 393, healthy = 713, total n = 1106
#filter those with phq2
with_phq2 <- dep_and_healthy_groups_for_ICD_analysis %>% filter(!is.na(PHQ.2))
#lm
fascicle_lm <- lapply(fascicle_names, function(x)
{
lm(substitute(i ~ PHQ.2, list(i = as.name(x))), data = with_phq2)
})
names(fascicle_lm) <- fascicle_names
#anova
fascicle_anova <- lapply(fascicle_lm, anova)
#fdr corrected
fascicle_anova_fdr <- fdr_anova_generic(fascicle_anova, 1)## p_anova
## AF_L 0.077464744
## AF_R 0.317066928
## C_FPH_L 0.816835799
## C_FPH_R 0.124243674
## C_FP_L 0.135807767
## C_FP_R 0.030943489
## C_PH_L 0.148740215
## C_PHP_L 0.045996122
## C_PHP_R 0.777904258
## C_PH_R 0.326428204
## C_R_L 0.297117359
## C_R_R 0.086061797
## EMC_L 0.090320659
## EMC_R 0.086891518
## FAT_L 0.035850802
## FAT_R 0.016370319
## IFOF_L 0.495097661
## IFOF_R 0.498219296
## ILF_L 0.550423607
## ILF_R 0.416295726
## MdLF_L 0.046911349
## MdLF_R 0.307883644
## PAT_L 0.348306882
## PAT_R 0.309623104
## SLF1_L 0.087417587
## SLF1_R 0.079839286
## SLF2_L 0.001709903
## SLF2_R 0.082919521
## SLF3_L 0.222574593
## SLF3_R 0.064252758
## UF_L 0.652324041
## UF_R 0.096690807
## VOF_L 0.074897166
## VOF_R 0.130184528
## CB_L 0.566187111
## CB_R 0.542767030
## ICP_L 0.791234248
## ICP_R 0.522728989
## MCP 0.406619100
## SCP 0.098839527
## V 0.682399418
## CNIII_L 0.543320701
## CNIII_R 0.541827308
## CNII_L 0.422969425
## CNII_R 0.162978091
## CNVIII_L 0.513909388
## CNVIII_R 0.046947968
## CNVII_L 0.759538768
## CNVII_R 0.797578096
## CNV_L 0.293650792
## CNV_R 0.431234510
## AR_L 0.931057184
## AR_R 0.329108835
## CBT_L 0.390943419
## CBT_R 0.073514371
## CPT_F_L 0.095896727
## CPT_F_R 0.027097044
## CPT_O_L 0.606585435
## CPT_O_R 0.911584104
## CPT_P_L 0.059734213
## CPT_P_R 0.496264301
## CS_A_L 0.548950787
## CS_A_R 0.062251050
## CS_P_L 0.225513477
## CS_P_R 0.772292622
## CS_S_L 0.191097895
## CS_S_R 0.062700715
## CST_L 0.097314217
## CST_R 0.353997355
## DRTT_L 0.045209052
## DRTT_R 0.061259641
## F_L 0.805901820
## F_R 0.330011771
## ML_L 0.084205265
## ML_R 0.130826349
## OR_L 0.140329561
## OR_R 0.048289914
## RST_L 0.005419655
## RST_R 0.009505794
## TR_A_L 0.310521082
## TR_A_R 0.042052804
## TR_P_L 0.424787995
## TR_P_R 0.948570297
## TR_S_L 0.315685651
## TR_S_R 0.113319889
## AC 0.361088235
## CC 0.775713382print(fascicle_anova_fdr)## [1] component p_FDR_corr
## <0 rows> (or 0-length row.names)#fdr corrected only association cortex
fascicle_anova_w_fiber_mapping <- fascicle_anova[which(fascicle_bundle_mapping$name_vector == "association")]
fascicle_anova_fdr_association <- fdr_anova_generic(fascicle_anova_w_fiber_mapping,1)## p_anova
## AF_L 0.077464744
## AF_R 0.317066928
## C_FPH_L 0.816835799
## C_FPH_R 0.124243674
## C_FP_L 0.135807767
## C_FP_R 0.030943489
## C_PH_L 0.148740215
## C_PHP_L 0.045996122
## C_PHP_R 0.777904258
## C_PH_R 0.326428204
## C_R_L 0.297117359
## C_R_R 0.086061797
## EMC_L 0.090320659
## EMC_R 0.086891518
## FAT_L 0.035850802
## FAT_R 0.016370319
## IFOF_L 0.495097661
## IFOF_R 0.498219296
## ILF_L 0.550423607
## ILF_R 0.416295726
## MdLF_L 0.046911349
## MdLF_R 0.307883644
## PAT_L 0.348306882
## PAT_R 0.309623104
## SLF1_L 0.087417587
## SLF1_R 0.079839286
## SLF2_L 0.001709903
## SLF2_R 0.082919521
## SLF3_L 0.222574593
## SLF3_R 0.064252758
## UF_L 0.652324041
## UF_R 0.096690807
## VOF_L 0.074897166
## VOF_R 0.130184528print(fascicle_anova_fdr_association)## [1] component p_FDR_corr
## <0 rows> (or 0-length row.names)#uncorrected
fascicle_anova_p <- sapply(fascicle_anova, function(v) v$"Pr(>F)"[1])
fascicle_anova_p05_unc <- as.data.frame(fascicle_anova_p[fascicle_anova_p < 0.05])
print(fascicle_anova_p05_unc)## fascicle_anova_p[fascicle_anova_p < 0.05]
## C_FP_R 0.030943489
## C_PHP_L 0.045996122
## FAT_L 0.035850802
## FAT_R 0.016370319
## MdLF_L 0.046911349
## SLF2_L 0.001709903
## CNVIII_R 0.046947968
## CPT_F_R 0.027097044
## DRTT_L 0.045209052
## OR_R 0.048289914
## RST_L 0.005419655
## RST_R 0.009505794
## TR_A_R 0.042052804#visreg
#sapply(fascicle_lm, visreg)#isolate out only the depressed group and healthy group
dep_groups_for_ICD_analysis <- df_demo_and_fascicles[df_demo_and_fascicles$depGroupVar == 1,] #n depressed = 393, healthy = 713, total n = 1106
#filter those with phq2
with_phq2_dep <- dep_groups_for_ICD_analysis %>% filter(!is.na(PHQ.2) & PHQ.2 != 0) %>% filter(depGroupVar == 1)
#lm
fascicle_lm <- lapply(fascicle_names, function(x)
{
lm(substitute(i ~ PHQ.2, list(i = as.name(x))), data = with_phq2_dep)
})
names(fascicle_lm) <- fascicle_names
#anova
fascicle_anova <- lapply(fascicle_lm, anova)
#fdr corrected
fascicle_anova_fdr <- fdr_anova_generic(fascicle_anova, 1)## p_anova
## AF_L 0.483383605
## AF_R 0.482945508
## C_FPH_L 0.851708401
## C_FPH_R 0.068372041
## C_FP_L 0.719471943
## C_FP_R 0.242564839
## C_PH_L NaN
## C_PHP_L 0.141823242
## C_PHP_R 0.926294855
## C_PH_R 0.088300710
## C_R_L 0.068372041
## C_R_R 0.476971573
## EMC_L 0.658802646
## EMC_R 0.479447560
## FAT_L 0.838264222
## FAT_R 0.127600602
## IFOF_L 0.486309746
## IFOF_R 0.421909387
## ILF_L 0.973501398
## ILF_R 0.377652612
## MdLF_L 0.814790807
## MdLF_R 0.271402985
## PAT_L 0.506476645
## PAT_R 0.166894038
## SLF1_L 0.911406435
## SLF1_R 0.338021093
## SLF2_L 0.154855824
## SLF2_R 0.007041197
## SLF3_L 0.909371726
## SLF3_R 0.259110171
## UF_L 0.673858183
## UF_R 0.584198730
## VOF_L 0.009084999
## VOF_R 0.048145161
## CB_L NaN
## CB_R NaN
## ICP_L 0.851708401
## ICP_R 0.851708401
## MCP 0.851708401
## SCP 0.602403333
## V NaN
## CNIII_L NaN
## CNIII_R NaN
## CNII_L NaN
## CNII_R 0.557002885
## CNVIII_L NaN
## CNVIII_R 0.851708401
## CNVII_L NaN
## CNVII_R NaN
## CNV_L 0.851708401
## CNV_R NaN
## AR_L 0.953910347
## AR_R 0.609038346
## CBT_L 0.285687519
## CBT_R 0.367527244
## CPT_F_L 0.328624246
## CPT_F_R 0.156025570
## CPT_O_L 0.819070873
## CPT_O_R 0.989074818
## CPT_P_L 0.959283218
## CPT_P_R 0.549980016
## CS_A_L 0.572216088
## CS_A_R 0.072371682
## CS_P_L 0.764778316
## CS_P_R 0.569741081
## CS_S_L 0.630727404
## CS_S_R 0.169331419
## CST_L 0.835455381
## CST_R 0.329056372
## DRTT_L 0.271615258
## DRTT_R 0.042071840
## F_L 0.324223773
## F_R 0.097360242
## ML_L 0.589062398
## ML_R 0.015167059
## OR_L 0.219578859
## OR_R 0.125087202
## RST_L 0.138098774
## RST_R 0.020030499
## TR_A_L 0.403952193
## TR_A_R 0.021116772
## TR_P_L 0.917561102
## TR_P_R 0.963951534
## TR_S_L 0.713223938
## TR_S_R 0.344277890
## AC 0.069156063
## CC 0.488871749print(fascicle_anova_fdr)## component p_FDR_corr
## 1 <NA> <NA>
## 2 <NA> <NA>
## 3 <NA> <NA>
## 4 <NA> <NA>
## 5 <NA> <NA>
## 6 <NA> <NA>
## 7 <NA> <NA>
## 8 <NA> <NA>
## 9 <NA> <NA>
## 10 <NA> <NA>
## 11 <NA> <NA>#fdr corrected only association cortex
fascicle_anova_w_fiber_mapping <- fascicle_anova[which(fascicle_bundle_mapping$name_vector == "association")]
fascicle_anova_fdr_association <- fdr_anova_generic(fascicle_anova_w_fiber_mapping,1)## p_anova
## AF_L 0.483383605
## AF_R 0.482945508
## C_FPH_L 0.851708401
## C_FPH_R 0.068372041
## C_FP_L 0.719471943
## C_FP_R 0.242564839
## C_PH_L NaN
## C_PHP_L 0.141823242
## C_PHP_R 0.926294855
## C_PH_R 0.088300710
## C_R_L 0.068372041
## C_R_R 0.476971573
## EMC_L 0.658802646
## EMC_R 0.479447560
## FAT_L 0.838264222
## FAT_R 0.127600602
## IFOF_L 0.486309746
## IFOF_R 0.421909387
## ILF_L 0.973501398
## ILF_R 0.377652612
## MdLF_L 0.814790807
## MdLF_R 0.271402985
## PAT_L 0.506476645
## PAT_R 0.166894038
## SLF1_L 0.911406435
## SLF1_R 0.338021093
## SLF2_L 0.154855824
## SLF2_R 0.007041197
## SLF3_L 0.909371726
## SLF3_R 0.259110171
## UF_L 0.673858183
## UF_R 0.584198730
## VOF_L 0.009084999
## VOF_R 0.048145161print(fascicle_anova_fdr_association)## component p_FDR_corr
## 1 <NA> <NA>#uncorrected
fascicle_anova_p <- sapply(fascicle_anova, function(v) v$"Pr(>F)"[1])
fascicle_anova_p05_unc <- as.data.frame(fascicle_anova_p[fascicle_anova_p < 0.05])
print(fascicle_anova_p05_unc)## fascicle_anova_p[fascicle_anova_p < 0.05]
## 1 NA
## 2 0.007041197
## 3 0.009084999
## 4 0.048145161
## 5 NA
## 6 NA
## 7 NA
## 8 NA
## 9 NA
## 10 NA
## 11 NA
## 12 NA
## 13 NA
## 14 NA
## 15 0.042071840
## 16 0.015167059
## 17 0.020030499
## 18 0.021116772#visreg
#sapply(fascicle_lm, visreg)#isolate out only the depressed group and healthy group
dep_and_healthy_groups_for_ICD_analysis <- df_demo_and_fascicles[df_demo_and_fascicles$depGroupVar != 0,] #n depressed = 393, healthy = 713, total n = 1106
#filter those with phq2
with_phq9 <- dep_and_healthy_groups_for_ICD_analysis %>% filter(!is.na(PHQ.9))
#lm
fascicle_lm <- lapply(fascicle_names, function(x)
{
lm(substitute(i ~ PHQ.9, list(i = as.name(x))), data = with_phq9)
})
names(fascicle_lm) <- fascicle_names
#anova
fascicle_anova <- lapply(fascicle_lm, anova)
#fdr corrected
fascicle_anova_fdr <- fdr_anova_generic(fascicle_anova, 1)## p_anova
## AF_L 0.707990735
## AF_R 0.477814266
## C_FPH_L NaN
## C_FPH_R NaN
## C_FP_L 0.766121519
## C_FP_R 0.064792487
## C_PH_L NaN
## C_PHP_L 0.252908243
## C_PHP_R 0.709741992
## C_PH_R 0.009755309
## C_R_L 0.167591862
## C_R_R 0.098674053
## EMC_L 0.062785768
## EMC_R 0.598626499
## FAT_L 0.840134923
## FAT_R 0.615654520
## IFOF_L 0.001934431
## IFOF_R 0.925611046
## ILF_L 0.013221750
## ILF_R 0.800775172
## MdLF_L 0.190371407
## MdLF_R 0.982299309
## PAT_L 0.458654730
## PAT_R 0.454952752
## SLF1_L 0.747379443
## SLF1_R 0.264778653
## SLF2_L 0.670378026
## SLF2_R 0.440822863
## SLF3_L 0.797000652
## SLF3_R 0.254086319
## UF_L 0.847264238
## UF_R 0.368790280
## VOF_L 0.007383087
## VOF_R 0.270311447
## CB_L NaN
## CB_R 0.098674053
## ICP_L NaN
## ICP_R NaN
## MCP NaN
## SCP 0.349095868
## V 0.098674053
## CNIII_L NaN
## CNIII_R NaN
## CNII_L NaN
## CNII_R 0.200712922
## CNVIII_L NaN
## CNVIII_R NaN
## CNVII_L NaN
## CNVII_R NaN
## CNV_L 0.493620715
## CNV_R 0.571891059
## AR_L 0.378629429
## AR_R 0.049599908
## CBT_L 0.281190669
## CBT_R 0.220807058
## CPT_F_L 0.583114313
## CPT_F_R 0.995641671
## CPT_O_L 0.711454101
## CPT_O_R 0.038456873
## CPT_P_L 0.098936218
## CPT_P_R 0.584079560
## CS_A_L 0.177874046
## CS_A_R 0.672969904
## CS_P_L 0.235216217
## CS_P_R 0.028439943
## CS_S_L 0.238002149
## CS_S_R 0.982985505
## CST_L 0.410616700
## CST_R 0.506177882
## DRTT_L 0.307671460
## DRTT_R 0.574700336
## F_L 0.583718784
## F_R 0.944225981
## ML_L 0.048198827
## ML_R 0.262371675
## OR_L 0.216791942
## OR_R 0.531013236
## RST_L 0.143211474
## RST_R 0.615046508
## TR_A_L 0.144652698
## TR_A_R 0.781587639
## TR_P_L 0.493783599
## TR_P_R 0.268291739
## TR_S_L 0.249354201
## TR_S_R 0.628426611
## AC 0.461065637
## CC 0.688285315print(fascicle_anova_fdr)## component p_FDR_corr
## 1 <NA> <NA>
## 2 <NA> <NA>
## 3 <NA> <NA>
## 4 <NA> <NA>
## 5 <NA> <NA>
## 6 <NA> <NA>
## 7 <NA> <NA>
## 8 <NA> <NA>
## 9 <NA> <NA>
## 10 <NA> <NA>
## 11 <NA> <NA>
## 12 <NA> <NA>
## 13 <NA> <NA>
## 14 <NA> <NA>#fdr corrected only association cortex
fascicle_anova_w_fiber_mapping <- fascicle_anova[which(fascicle_bundle_mapping$name_vector == "association")]
fascicle_anova_fdr_association <- fdr_anova_generic(fascicle_anova_w_fiber_mapping,1)## p_anova
## AF_L 0.707990735
## AF_R 0.477814266
## C_FPH_L NaN
## C_FPH_R NaN
## C_FP_L 0.766121519
## C_FP_R 0.064792487
## C_PH_L NaN
## C_PHP_L 0.252908243
## C_PHP_R 0.709741992
## C_PH_R 0.009755309
## C_R_L 0.167591862
## C_R_R 0.098674053
## EMC_L 0.062785768
## EMC_R 0.598626499
## FAT_L 0.840134923
## FAT_R 0.615654520
## IFOF_L 0.001934431
## IFOF_R 0.925611046
## ILF_L 0.013221750
## ILF_R 0.800775172
## MdLF_L 0.190371407
## MdLF_R 0.982299309
## PAT_L 0.458654730
## PAT_R 0.454952752
## SLF1_L 0.747379443
## SLF1_R 0.264778653
## SLF2_L 0.670378026
## SLF2_R 0.440822863
## SLF3_L 0.797000652
## SLF3_R 0.254086319
## UF_L 0.847264238
## UF_R 0.368790280
## VOF_L 0.007383087
## VOF_R 0.270311447print(fascicle_anova_fdr_association)## component p_FDR_corr
## 1 <NA> <NA>
## 2 <NA> <NA>
## 3 <NA> <NA>#uncorrected
fascicle_anova_p <- sapply(fascicle_anova, function(v) v$"Pr(>F)"[1])
fascicle_anova_p05_unc <- as.data.frame(fascicle_anova_p[fascicle_anova_p < 0.05])
print(fascicle_anova_p05_unc)## fascicle_anova_p[fascicle_anova_p < 0.05]
## 1 NA
## 2 NA
## 3 NA
## 4 0.009755309
## 5 0.001934431
## 6 0.013221750
## 7 0.007383087
## 8 NA
## 9 NA
## 10 NA
## 11 NA
## 12 NA
## 13 NA
## 14 NA
## 15 NA
## 16 NA
## 17 NA
## 18 NA
## 19 0.049599908
## 20 0.038456873
## 21 0.028439943
## 22 0.048198827#visreg
#sapply(fascicle_lm, visreg)#isolate out only the depressed group and healthy group
dep_groups_for_ICD_analysis <- df_demo_and_fascicles[df_demo_and_fascicles$depGroupVar == 1,] #n depressed = 393, healthy = 713, total n = 1106
#filter those with phq2
with_phq9_dep <- dep_groups_for_ICD_analysis %>% filter(!is.na(PHQ.9) & PHQ.9 != 0)
#lm
fascicle_lm <- lapply(fascicle_names, function(x)
{
lm(substitute(i ~ PHQ.9, list(i = as.name(x))), data = with_phq9_dep)
})
names(fascicle_lm) <- fascicle_names
#anova
fascicle_anova <- lapply(fascicle_lm, anova)
#fdr corrected
fascicle_anova_fdr <- fdr_anova_generic(fascicle_anova, 1)## p_anova
## AF_L 5.739085e-01
## AF_R 9.184613e-01
## C_FPH_L NaN
## C_FPH_R NaN
## C_FP_L 9.510758e-01
## C_FP_R 2.033026e-01
## C_PH_L NaN
## C_PHP_L 9.510758e-01
## C_PHP_R 8.835641e-01
## C_PH_R NaN
## C_R_L 5.980156e-02
## C_R_R NaN
## EMC_L 1.666289e-01
## EMC_R 1.915392e-01
## FAT_L 1.752849e-01
## FAT_R 8.330424e-01
## IFOF_L 1.915317e-02
## IFOF_R 9.246414e-01
## ILF_L 2.424471e-01
## ILF_R 7.427261e-01
## MdLF_L 1.965325e-01
## MdLF_R 3.077952e-01
## PAT_L 9.324089e-01
## PAT_R 9.046469e-01
## SLF1_L 4.405830e-01
## SLF1_R 4.123149e-01
## SLF2_L 7.294137e-01
## SLF2_R 9.757208e-01
## SLF3_L 9.226955e-01
## SLF3_R 9.339746e-01
## UF_L 8.727534e-01
## UF_R 3.892042e-01
## VOF_L 8.009238e-02
## VOF_R 5.077380e-01
## CB_L NaN
## CB_R NaN
## ICP_L NaN
## ICP_R NaN
## MCP NaN
## SCP 3.442412e-01
## V NaN
## CNIII_L NaN
## CNIII_R NaN
## CNII_L NaN
## CNII_R 1.545463e-01
## CNVIII_L NaN
## CNVIII_R NaN
## CNVII_L NaN
## CNVII_R NaN
## CNV_L 1.530614e-01
## CNV_R 3.185195e-01
## AR_L 6.661055e-01
## AR_R 1.401860e-01
## CBT_L 2.672927e-01
## CBT_R 7.939461e-01
## CPT_F_L 1.362785e-01
## CPT_F_R 5.405124e-01
## CPT_O_L 7.495423e-01
## CPT_O_R 1.575150e-02
## CPT_P_L 3.379268e-01
## CPT_P_R 1.762960e-03
## CS_A_L 2.993676e-01
## CS_A_R 3.987155e-01
## CS_P_L 4.261475e-01
## CS_P_R 1.278336e-02
## CS_S_L 1.328850e-01
## CS_S_R 4.039941e-01
## CST_L 5.146108e-01
## CST_R 6.468565e-01
## DRTT_L 2.549529e-01
## DRTT_R 3.992747e-01
## F_L 5.239275e-01
## F_R 4.315353e-01
## ML_L 1.137454e-01
## ML_R 3.117837e-06
## OR_L 7.735658e-01
## OR_R 9.978432e-01
## RST_L 4.570528e-02
## RST_R 8.340034e-01
## TR_A_L 3.773566e-01
## TR_A_R 4.759332e-01
## TR_P_L 8.631589e-01
## TR_P_R 2.928581e-01
## TR_S_L 1.184173e-01
## TR_S_R 2.850988e-01
## AC 6.919131e-01
## CC 2.027914e-01print(fascicle_anova_fdr)## component p_FDR_corr
## 1 <NA> <NA>
## 2 <NA> <NA>
## 3 <NA> <NA>
## 4 <NA> <NA>
## 5 <NA> <NA>
## 6 <NA> <NA>
## 7 <NA> <NA>
## 8 <NA> <NA>
## 9 <NA> <NA>
## 10 <NA> <NA>
## 11 <NA> <NA>
## 12 <NA> <NA>
## 13 <NA> <NA>
## 14 <NA> <NA>
## 15 <NA> <NA>
## 16 <NA> <NA>
## 17 <NA> <NA>
## 18 <NA> <NA>
## 19 ML_R 0#fdr corrected only association cortex
fascicle_anova_w_fiber_mapping <- fascicle_anova[which(fascicle_bundle_mapping$name_vector == "association")]
fascicle_anova_fdr_association <- fdr_anova_generic(fascicle_anova_w_fiber_mapping,1)## p_anova
## AF_L 0.57390846
## AF_R 0.91846134
## C_FPH_L NaN
## C_FPH_R NaN
## C_FP_L 0.95107576
## C_FP_R 0.20330264
## C_PH_L NaN
## C_PHP_L 0.95107576
## C_PHP_R 0.88356412
## C_PH_R NaN
## C_R_L 0.05980156
## C_R_R NaN
## EMC_L 0.16662886
## EMC_R 0.19153922
## FAT_L 0.17528494
## FAT_R 0.83304241
## IFOF_L 0.01915317
## IFOF_R 0.92464141
## ILF_L 0.24244707
## ILF_R 0.74272608
## MdLF_L 0.19653253
## MdLF_R 0.30779517
## PAT_L 0.93240885
## PAT_R 0.90464690
## SLF1_L 0.44058301
## SLF1_R 0.41231490
## SLF2_L 0.72941372
## SLF2_R 0.97572085
## SLF3_L 0.92269552
## SLF3_R 0.93397464
## UF_L 0.87275345
## UF_R 0.38920419
## VOF_L 0.08009238
## VOF_R 0.50773800print(fascicle_anova_fdr_association)## component p_FDR_corr
## 1 <NA> <NA>
## 2 <NA> <NA>
## 3 <NA> <NA>
## 4 <NA> <NA>
## 5 <NA> <NA>#uncorrected
fascicle_anova_p <- sapply(fascicle_anova, function(v) v$"Pr(>F)"[1])
fascicle_anova_p05_unc <- as.data.frame(fascicle_anova_p[fascicle_anova_p < 0.05])
print(fascicle_anova_p05_unc)## fascicle_anova_p[fascicle_anova_p < 0.05]
## 1 NA
## 2 NA
## 3 NA
## 4 NA
## 5 NA
## 6 1.915317e-02
## 7 NA
## 8 NA
## 9 NA
## 10 NA
## 11 NA
## 12 NA
## 13 NA
## 14 NA
## 15 NA
## 16 NA
## 17 NA
## 18 NA
## 19 NA
## 20 1.575150e-02
## 21 1.762960e-03
## 22 1.278336e-02
## 23 3.117837e-06
## 24 4.570528e-02#visreg
#sapply(fascicle_lm, visreg)##### repeat above analysis but use only the first instance of each EMPI, sorted by date
df_unique_empi <- dep_and_healthy_groups_for_ICD_analysis %>%
group_by(EMPI) %>%
arrange(EXAM_DATE) %>%
slice(1) %>%
ungroup() #n = 300, 115 depressed, 185 healthy
#lm
fascicle_lm_unique <- lapply(fascicle_names, function(x)
{
lm(substitute(i ~ depGroupVar, list(i = as.name(x))), data = df_unique_empi)
})
names(fascicle_lm_unique) <- fascicle_names
#anova
fascicle_anova_unique <- lapply(fascicle_lm_unique, anova)
#fdr corrected
fascicle_anova_unique_fdr <- fdr_anova_generic(fascicle_anova_unique, 1)## p_anova
## AF_L 0.02567250
## AF_R 0.04770448
## C_FPH_L 0.52391294
## C_FPH_R 0.65399543
## C_FP_L 0.48257224
## C_FP_R 0.45804086
## C_PH_L 0.01296162
## C_PHP_L 0.89250446
## C_PHP_R 0.69018950
## C_PH_R 0.46121161
## C_R_L 0.97178852
## C_R_R 0.21083969
## EMC_L 0.31552969
## EMC_R 0.79906299
## FAT_L 0.14705546
## FAT_R 0.07693380
## IFOF_L 0.08459977
## IFOF_R 0.33245803
## ILF_L 0.02549913
## ILF_R 0.16956127
## MdLF_L 0.37358728
## MdLF_R 0.24894845
## PAT_L 0.06478979
## PAT_R 0.01735310
## SLF1_L 0.03644751
## SLF1_R 0.59627268
## SLF2_L 0.23908639
## SLF2_R 0.06233292
## SLF3_L 0.18713491
## SLF3_R 0.17736128
## UF_L 0.67930840
## UF_R 0.57479717
## VOF_L 0.56658390
## VOF_R 0.01189975
## CB_L 0.86501258
## CB_R 0.81213034
## ICP_L 0.99405802
## ICP_R 0.27810364
## MCP 0.54939427
## SCP 0.04967510
## V 0.50289423
## CNIII_L 0.56195527
## CNIII_R 0.46653398
## CNII_L 0.09357141
## CNII_R 0.34501658
## CNVIII_L 0.68381315
## CNVIII_R 0.27648526
## CNVII_L 0.43136407
## CNVII_R 0.43136407
## CNV_L 0.38363115
## CNV_R 0.32577122
## AR_L 0.08576154
## AR_R 0.06881451
## CBT_L 0.13009678
## CBT_R 0.10177583
## CPT_F_L 0.16822470
## CPT_F_R 0.03750849
## CPT_O_L 0.06952521
## CPT_O_R 0.34891233
## CPT_P_L 0.56189636
## CPT_P_R 0.09021897
## CS_A_L 0.32434188
## CS_A_R 0.16230369
## CS_P_L 0.34417997
## CS_P_R 0.44656372
## CS_S_L 0.14708780
## CS_S_R 0.12134611
## CST_L 0.38231082
## CST_R 0.09567975
## DRTT_L 0.12146445
## DRTT_R 0.01457703
## F_L 0.02068663
## F_R 0.19322091
## ML_L 0.68628017
## ML_R 0.22670168
## OR_L 0.29395359
## OR_R 0.19239239
## RST_L 0.07766356
## RST_R 0.03199621
## TR_A_L 0.26505645
## TR_A_R 0.01705549
## TR_P_L 0.15036940
## TR_P_R 0.10361940
## TR_S_L 0.18944704
## TR_S_R 0.02805414
## AC 0.21650960
## CC 0.16499416print(fascicle_anova_unique_fdr)#anova values < 0.05, uncorrected## [1] component p_FDR_corr
## <0 rows> (or 0-length row.names)#fdr corrected only association cortex
fascicle_anova_w_fiber_mapping <- fascicle_anova_unique[which(fascicle_bundle_mapping$name_vector == "association")]
fascicle_anova_fdr_association <- fdr_anova_generic(fascicle_anova_w_fiber_mapping,1)## p_anova
## AF_L 0.02567250
## AF_R 0.04770448
## C_FPH_L 0.52391294
## C_FPH_R 0.65399543
## C_FP_L 0.48257224
## C_FP_R 0.45804086
## C_PH_L 0.01296162
## C_PHP_L 0.89250446
## C_PHP_R 0.69018950
## C_PH_R 0.46121161
## C_R_L 0.97178852
## C_R_R 0.21083969
## EMC_L 0.31552969
## EMC_R 0.79906299
## FAT_L 0.14705546
## FAT_R 0.07693380
## IFOF_L 0.08459977
## IFOF_R 0.33245803
## ILF_L 0.02549913
## ILF_R 0.16956127
## MdLF_L 0.37358728
## MdLF_R 0.24894845
## PAT_L 0.06478979
## PAT_R 0.01735310
## SLF1_L 0.03644751
## SLF1_R 0.59627268
## SLF2_L 0.23908639
## SLF2_R 0.06233292
## SLF3_L 0.18713491
## SLF3_R 0.17736128
## UF_L 0.67930840
## UF_R 0.57479717
## VOF_L 0.56658390
## VOF_R 0.01189975print(fascicle_anova_fdr_association)## [1] component p_FDR_corr
## <0 rows> (or 0-length row.names)#uncorrected
fascicle_anova_unique_p <- sapply(fascicle_anova_unique, function(v) v$"Pr(>F)"[1])
fascicle_anova_unique_p05_unc <- as.data.frame(fascicle_anova_unique_p[fascicle_anova_unique_p < 0.05])
print(fascicle_anova_unique_p05_unc)## fascicle_anova_unique_p[fascicle_anova_unique_p < 0.05]
## AF_L 0.02567250
## AF_R 0.04770448
## C_PH_L 0.01296162
## ILF_L 0.02549913
## PAT_R 0.01735310
## SLF1_L 0.03644751
## VOF_R 0.01189975
## SCP 0.04967510
## CPT_F_R 0.03750849
## DRTT_R 0.01457703
## F_L 0.02068663
## RST_R 0.03199621
## TR_A_R 0.01705549
## TR_S_R 0.02805414##### repeat above analysis but use only the first instance of each EMPI, sorted by date
df_mm <-df_demo_and_fascicles[df_demo_and_fascicles$depGroupVar != 0,]
#lm
fascicle_lm_mm <- lapply(fascicle_names, function(x)
{
lmerTest::lmer(substitute(i ~ depGroupVar + (1 | EMPI), list(i = as.name(x))), data = df_mm)
})
names(fascicle_lm_mm) <- fascicle_names
#anova
fascicle_anova_mm <- lapply(fascicle_lm_mm, anova)
#fdr corrected
fascicle_anova_mm_fdr <- fdr_anova_generic(fascicle_anova_mm, 1)## p_anova
## AF_L 0.07356432
## AF_R 0.14759628
## C_FPH_L 0.76125615
## C_FPH_R 0.95881596
## C_FP_L 0.56708141
## C_FP_R 0.89667874
## C_PH_L 0.04145253
## C_PHP_L 0.82066826
## C_PHP_R 0.53029048
## C_PH_R 0.98486035
## C_R_L 0.28392320
## C_R_R 0.39646734
## EMC_L 0.28985250
## EMC_R 0.81932484
## FAT_L 0.26949553
## FAT_R 0.18141800
## IFOF_L 0.14437402
## IFOF_R 0.36790993
## ILF_L 0.05360218
## ILF_R 0.24513872
## MdLF_L 0.31625991
## MdLF_R 0.32429030
## PAT_L 0.10603016
## PAT_R 0.03268715
## SLF1_L 0.06965037
## SLF1_R 0.93709161
## SLF2_L 0.33911788
## SLF2_R 0.19157412
## SLF3_L 0.35271618
## SLF3_R 0.35359715
## UF_L 0.51120672
## UF_R 0.63279931
## VOF_L 0.75821032
## VOF_R 0.02797395
## CB_L 0.76304320
## CB_R 0.82906592
## ICP_L 0.75185593
## ICP_R 0.89669523
## MCP 0.90395851
## SCP 0.43859301
## V 0.92634098
## CNIII_L 0.79479737
## CNIII_R 0.86639305
## CNII_L 0.92674516
## CNII_R 0.43809445
## CNVIII_L 0.39106717
## CNVIII_R 0.70603078
## CNVII_L 0.33483655
## CNVII_R 0.44050074
## CNV_L 0.43970798
## CNV_R 0.09900285
## AR_L 0.15608222
## AR_R 0.09959434
## CBT_L 0.22163179
## CBT_R 0.25793906
## CPT_F_L 0.24542545
## CPT_F_R 0.09106739
## CPT_O_L 0.14936215
## CPT_O_R 0.45444091
## CPT_P_L 0.58535096
## CPT_P_R 0.13278535
## CS_A_L 0.23769121
## CS_A_R 0.30888866
## CS_P_L 0.40792019
## CS_P_R 0.64678285
## CS_S_L 0.21563567
## CS_S_R 0.20340358
## CST_L 0.53562485
## CST_R 0.15656062
## DRTT_L 0.32079032
## DRTT_R 0.04505424
## F_L 0.08668841
## F_R 0.53201429
## ML_L 0.91075218
## ML_R 0.11544616
## OR_L 0.46145351
## OR_R 0.28228296
## RST_L 0.16426409
## RST_R 0.07758430
## TR_A_L 0.16992237
## TR_A_R 0.05569790
## TR_P_L 0.19265851
## TR_P_R 0.14234638
## TR_S_L 0.29842169
## TR_S_R 0.05127563
## AC 0.38475089
## CC 0.29738221print(fascicle_anova_mm_fdr)#anova values < 0.05, uncorrected## [1] component p_FDR_corr
## <0 rows> (or 0-length row.names)#fdr corrected only association cortex
fascicle_anova_w_fiber_mapping <- fascicle_anova_mm[which(fascicle_bundle_mapping$name_vector == "association")]
fascicle_anova_fdr_association <- fdr_anova_generic(fascicle_anova_w_fiber_mapping,1)## p_anova
## AF_L 0.07356432
## AF_R 0.14759628
## C_FPH_L 0.76125615
## C_FPH_R 0.95881596
## C_FP_L 0.56708141
## C_FP_R 0.89667874
## C_PH_L 0.04145253
## C_PHP_L 0.82066826
## C_PHP_R 0.53029048
## C_PH_R 0.98486035
## C_R_L 0.28392320
## C_R_R 0.39646734
## EMC_L 0.28985250
## EMC_R 0.81932484
## FAT_L 0.26949553
## FAT_R 0.18141800
## IFOF_L 0.14437402
## IFOF_R 0.36790993
## ILF_L 0.05360218
## ILF_R 0.24513872
## MdLF_L 0.31625991
## MdLF_R 0.32429030
## PAT_L 0.10603016
## PAT_R 0.03268715
## SLF1_L 0.06965037
## SLF1_R 0.93709161
## SLF2_L 0.33911788
## SLF2_R 0.19157412
## SLF3_L 0.35271618
## SLF3_R 0.35359715
## UF_L 0.51120672
## UF_R 0.63279931
## VOF_L 0.75821032
## VOF_R 0.02797395print(fascicle_anova_fdr_association)## [1] component p_FDR_corr
## <0 rows> (or 0-length row.names)#uncorrected
fascicle_anova_mm_p <- sapply(fascicle_anova_mm, function(v) v$"Pr(>F)"[1])
fascicle_anova_mm_p05_unc <- as.data.frame(fascicle_anova_mm_p[fascicle_anova_mm_p < 0.05])
print(fascicle_anova_mm_p05_unc)## fascicle_anova_mm_p[fascicle_anova_mm_p < 0.05]
## C_PH_L 0.04145253
## PAT_R 0.03268715
## VOF_R 0.02797395
## DRTT_R 0.04505424########### age effect whole group
#isolate out only the depressed group and healthy group
dep_and_healthy_groups_for_ICD_analysis <- df_demo_and_fascicles[df_demo_and_fascicles$depGroupVar != 0,] #n depressed = 393, healthy = 713, total n = 1106
#### age histo ###
hist(dep_and_healthy_groups_for_ICD_analysis$PAT_AGE_AT_EXAM,
main = paste0("Age In Good Mimosa Group : n = ", dim(dep_and_healthy_groups_for_ICD_analysis)[1]), xlab = "Age", breaks = 10)
#lm
fascicle_age_lm <- lapply(fascicle_names, function(x)
{
lm(substitute(i ~ PAT_AGE_AT_EXAM, list(i = as.name(x))), data = dep_and_healthy_groups_for_ICD_analysis)
})
names(fascicle_age_lm) <- fascicle_names
#anova
fascicle_age_anova <- lapply(fascicle_age_lm, anova)
#fdr corrected
fascicle_age_anova_fdr <- fdr_anova_generic(fascicle_age_anova, 1)## p_anova
## AF_L 7.675516e-12
## AF_R 5.055895e-05
## C_FPH_L 8.258035e-03
## C_FPH_R 8.227520e-01
## C_FP_L 3.004545e-05
## C_FP_R 2.885625e-02
## C_PH_L 5.130225e-01
## C_PHP_L 2.952220e-04
## C_PHP_R 6.970269e-04
## C_PH_R 5.207577e-04
## C_R_L 4.424656e-03
## C_R_R 2.867211e-04
## EMC_L 2.219735e-06
## EMC_R 7.113004e-04
## FAT_L 2.001993e-14
## FAT_R 4.136943e-14
## IFOF_L 5.954939e-05
## IFOF_R 3.623102e-03
## ILF_L 1.245812e-03
## ILF_R 1.801121e-01
## MdLF_L 4.548818e-04
## MdLF_R 6.685245e-01
## PAT_L 1.850682e-10
## PAT_R 6.070195e-10
## SLF1_L 3.972055e-06
## SLF1_R 3.492555e-02
## SLF2_L 1.953684e-10
## SLF2_R 1.329061e-08
## SLF3_L 2.087589e-23
## SLF3_R 7.297727e-18
## UF_L 9.780502e-17
## UF_R 8.532133e-27
## VOF_L 4.644074e-02
## VOF_R 7.913203e-05
## CB_L 3.272553e-01
## CB_R 5.570468e-01
## ICP_L 3.210947e-01
## ICP_R 1.986896e-02
## MCP 9.800711e-01
## SCP 2.922375e-07
## V 5.601201e-01
## CNIII_L 7.783680e-01
## CNIII_R 2.838869e-01
## CNII_L 9.202604e-01
## CNII_R 2.219778e-02
## CNVIII_L 3.354313e-03
## CNVIII_R 3.001256e-02
## CNVII_L 4.433311e-02
## CNVII_R 5.494444e-02
## CNV_L 1.661940e-02
## CNV_R 4.313178e-01
## AR_L 8.703386e-02
## AR_R 1.019340e-01
## CBT_L 3.296664e-10
## CBT_R 1.979954e-11
## CPT_F_L 1.388156e-10
## CPT_F_R 3.354360e-07
## CPT_O_L 3.067032e-02
## CPT_O_R 2.004826e-02
## CPT_P_L 1.837730e-01
## CPT_P_R 6.035796e-01
## CS_A_L 1.286181e-23
## CS_A_R 1.171133e-16
## CS_P_L 1.278186e-01
## CS_P_R 2.064962e-03
## CS_S_L 2.418965e-09
## CS_S_R 8.237142e-10
## CST_L 3.107629e-02
## CST_R 5.050816e-02
## DRTT_L 2.976425e-06
## DRTT_R 1.333224e-07
## F_L 8.434985e-01
## F_R 5.960281e-01
## ML_L 1.114892e-01
## ML_R 6.331454e-01
## OR_L 4.213239e-03
## OR_R 4.525175e-03
## RST_L 7.897668e-17
## RST_R 2.603912e-14
## TR_A_L 8.847621e-24
## TR_A_R 1.985093e-17
## TR_P_L 1.335421e-01
## TR_P_R 1.495308e-02
## TR_S_L 7.172037e-08
## TR_S_R 1.213817e-06
## AC 1.543925e-03
## CC 8.497713e-04print(fascicle_age_anova_fdr)## component p_FDR_corr
## 1 AF_L 0
## 2 AF_R 0
## 3 C_FPH_L 0.014
## 4 C_FP_L 0
## 5 C_FP_R 0.045
## 6 C_PHP_L 0.001
## 7 C_PHP_R 0.002
## 8 C_PH_R 0.001
## 9 C_R_L 0.008
## 10 C_R_R 0.001
## 11 EMC_L 0
## 12 EMC_R 0.002
## 13 FAT_L 0
## 14 FAT_R 0
## 15 IFOF_L 0
## 16 IFOF_R 0.007
## 17 ILF_L 0.003
## 18 MdLF_L 0.001
## 19 PAT_L 0
## 20 PAT_R 0
## 21 SLF1_L 0
## 22 SLF2_L 0
## 23 SLF2_R 0
## 24 SLF3_L 0
## 25 SLF3_R 0
## 26 UF_L 0
## 27 UF_R 0
## 28 VOF_R 0
## 29 ICP_R 0.032
## 30 SCP 0
## 31 CNII_R 0.035
## 32 CNVIII_L 0.006
## 33 CNVIII_R 0.046
## 34 CNV_L 0.028
## 35 CBT_L 0
## 36 CBT_R 0
## 37 CPT_F_L 0
## 38 CPT_F_R 0
## 39 CPT_O_L 0.046
## 40 CPT_O_R 0.032
## 41 CS_A_L 0
## 42 CS_A_R 0
## 43 CS_P_R 0.004
## 44 CS_S_L 0
## 45 CS_S_R 0
## 46 CST_L 0.046
## 47 DRTT_L 0
## 48 DRTT_R 0
## 49 OR_L 0.008
## 50 OR_R 0.008
## 51 RST_L 0
## 52 RST_R 0
## 53 TR_A_L 0
## 54 TR_A_R 0
## 55 TR_P_R 0.026
## 56 TR_S_L 0
## 57 TR_S_R 0
## 58 AC 0.003
## 59 CC 0.002#fdr corrected only association cortex
fascicle_anova_w_fiber_mapping <- fascicle_age_anova[which(fascicle_bundle_mapping$name_vector == "association")]
fascicle_anova_fdr_association <- fdr_anova_generic(fascicle_anova_w_fiber_mapping,1)## p_anova
## AF_L 7.675516e-12
## AF_R 5.055895e-05
## C_FPH_L 8.258035e-03
## C_FPH_R 8.227520e-01
## C_FP_L 3.004545e-05
## C_FP_R 2.885625e-02
## C_PH_L 5.130225e-01
## C_PHP_L 2.952220e-04
## C_PHP_R 6.970269e-04
## C_PH_R 5.207577e-04
## C_R_L 4.424656e-03
## C_R_R 2.867211e-04
## EMC_L 2.219735e-06
## EMC_R 7.113004e-04
## FAT_L 2.001993e-14
## FAT_R 4.136943e-14
## IFOF_L 5.954939e-05
## IFOF_R 3.623102e-03
## ILF_L 1.245812e-03
## ILF_R 1.801121e-01
## MdLF_L 4.548818e-04
## MdLF_R 6.685245e-01
## PAT_L 1.850682e-10
## PAT_R 6.070195e-10
## SLF1_L 3.972055e-06
## SLF1_R 3.492555e-02
## SLF2_L 1.953684e-10
## SLF2_R 1.329061e-08
## SLF3_L 2.087589e-23
## SLF3_R 7.297727e-18
## UF_L 9.780502e-17
## UF_R 8.532133e-27
## VOF_L 4.644074e-02
## VOF_R 7.913203e-05print(fascicle_anova_fdr_association)## component p_FDR_corr
## 1 AF_L 0
## 2 AF_R 0
## 3 C_FPH_L 0.01
## 4 C_FP_L 0
## 5 C_FP_R 0.035
## 6 C_PHP_L 0.001
## 7 C_PHP_R 0.001
## 8 C_PH_R 0.001
## 9 C_R_L 0.006
## 10 C_R_R 0.001
## 11 EMC_L 0
## 12 EMC_R 0.001
## 13 FAT_L 0
## 14 FAT_R 0
## 15 IFOF_L 0
## 16 IFOF_R 0.005
## 17 ILF_L 0.002
## 18 MdLF_L 0.001
## 19 PAT_L 0
## 20 PAT_R 0
## 21 SLF1_L 0
## 22 SLF1_R 0.041
## 23 SLF2_L 0
## 24 SLF2_R 0
## 25 SLF3_L 0
## 26 SLF3_R 0
## 27 UF_L 0
## 28 UF_R 0
## 29 VOF_R 0#visreg
#sapply(fascicle_age_lm,visreg)########### age effect whole group
#isolate out only the depressed group and healthy group
dep_and_healthy_groups_for_ICD_analysis <- df_demo_and_fascicles[df_demo_and_fascicles$depGroupVar != 0,] %>%
group_by(EMPI) %>%
arrange(EXAM_DATE) %>%
slice(1) %>%
ungroup()
#### age histo ###
hist(dep_and_healthy_groups_for_ICD_analysis$PAT_AGE_AT_EXAM,
main = paste0("Age In Good Mimosa Group : n = ", dim(dep_and_healthy_groups_for_ICD_analysis)[1]), xlab = "Age", breaks = 10)
#lm
fascicle_age_lm <- lapply(fascicle_names, function(x)
{
lm(substitute(i ~ PAT_AGE_AT_EXAM, list(i = as.name(x))), data = dep_and_healthy_groups_for_ICD_analysis)
})
names(fascicle_age_lm) <- fascicle_names
#anova
fascicle_age_anova <- lapply(fascicle_age_lm, anova)
#fdr corrected
fascicle_age_anova_fdr <- fdr_anova_generic(fascicle_age_anova, 1)## p_anova
## AF_L 6.573644e-04
## AF_R 7.941737e-03
## C_FPH_L 5.260873e-01
## C_FPH_R 8.534352e-01
## C_FP_L 2.306614e-01
## C_FP_R 4.598187e-01
## C_PH_L 4.850014e-01
## C_PHP_L 1.465915e-02
## C_PHP_R 6.652227e-02
## C_PH_R 2.413927e-02
## C_R_L 7.296683e-01
## C_R_R 6.648078e-01
## EMC_L 3.059857e-03
## EMC_R 2.424058e-02
## FAT_L 3.380303e-05
## FAT_R 1.280932e-05
## IFOF_L 2.376973e-03
## IFOF_R 1.966622e-02
## ILF_L 4.039340e-02
## ILF_R 2.395401e-01
## MdLF_L 1.779880e-02
## MdLF_R 2.126197e-01
## PAT_L 7.147095e-03
## PAT_R 7.713464e-03
## SLF1_L 4.057589e-02
## SLF1_R 5.057219e-01
## SLF2_L 3.214374e-03
## SLF2_R 3.787585e-04
## SLF3_L 1.407435e-05
## SLF3_R 5.539809e-07
## UF_L 1.205368e-04
## UF_R 4.476156e-09
## VOF_L 5.287183e-01
## VOF_R 5.392905e-02
## CB_L 1.669512e-01
## CB_R 3.184672e-01
## ICP_L 6.323200e-01
## ICP_R 2.890181e-01
## MCP 6.497121e-01
## SCP 2.806354e-02
## V 1.083601e-01
## CNIII_L 7.119083e-01
## CNIII_R 7.735391e-01
## CNII_L 6.800387e-01
## CNII_R 5.634112e-01
## CNVIII_L 2.475735e-01
## CNVIII_R 4.199777e-01
## CNVII_L 7.844143e-01
## CNVII_R 1.681371e-01
## CNV_L 2.705120e-01
## CNV_R 4.635739e-01
## AR_L 9.011116e-02
## AR_R 1.961557e-01
## CBT_L 1.004508e-02
## CBT_R 1.313179e-04
## CPT_F_L 3.278054e-04
## CPT_F_R 2.304298e-03
## CPT_O_L 2.780634e-02
## CPT_O_R 5.370068e-02
## CPT_P_L 1.389019e-01
## CPT_P_R 5.281063e-01
## CS_A_L 4.403996e-08
## CS_A_R 8.463728e-07
## CS_P_L 9.736486e-02
## CS_P_R 2.036162e-02
## CS_S_L 1.055134e-03
## CS_S_R 5.024270e-04
## CST_L 1.982846e-01
## CST_R 2.340062e-01
## DRTT_L 1.626660e-02
## DRTT_R 2.493031e-03
## F_L 5.701735e-01
## F_R 9.066545e-01
## ML_L 7.931481e-01
## ML_R 7.895195e-01
## OR_L 2.512786e-02
## OR_R 2.556019e-02
## RST_L 8.013498e-06
## RST_R 2.530021e-05
## TR_A_L 3.435051e-09
## TR_A_R 9.555448e-07
## TR_P_L 1.338022e-01
## TR_P_R 1.112951e-01
## TR_S_L 1.007147e-03
## TR_S_R 4.974446e-03
## AC 2.240035e-02
## CC 8.954545e-03print(fascicle_age_anova_fdr)## component p_FDR_corr
## 1 AF_L 0.003
## 2 AF_R 0.025
## 3 C_PHP_L 0.041
## 4 EMC_L 0.012
## 5 FAT_L 0
## 6 FAT_R 0
## 7 IFOF_L 0.01
## 8 MdLF_L 0.047
## 9 PAT_L 0.024
## 10 PAT_R 0.025
## 11 SLF2_L 0.012
## 12 SLF2_R 0.002
## 13 SLF3_L 0
## 14 SLF3_R 0
## 15 UF_L 0.001
## 16 UF_R 0
## 17 CBT_L 0.029
## 18 CBT_R 0.001
## 19 CPT_F_L 0.002
## 20 CPT_F_R 0.01
## 21 CS_A_L 0
## 22 CS_A_R 0
## 23 CS_S_L 0.005
## 24 CS_S_R 0.003
## 25 DRTT_L 0.044
## 26 DRTT_R 0.01
## 27 RST_L 0
## 28 RST_R 0
## 29 TR_A_L 0
## 30 TR_A_R 0
## 31 TR_S_L 0.005
## 32 TR_S_R 0.017
## 33 CC 0.027#fdr corrected only association cortex
fascicle_anova_w_fiber_mapping <- fascicle_age_anova[which(fascicle_bundle_mapping$name_vector == "association")]
fascicle_anova_fdr_association <- fdr_anova_generic(fascicle_anova_w_fiber_mapping,1)## p_anova
## AF_L 6.573644e-04
## AF_R 7.941737e-03
## C_FPH_L 5.260873e-01
## C_FPH_R 8.534352e-01
## C_FP_L 2.306614e-01
## C_FP_R 4.598187e-01
## C_PH_L 4.850014e-01
## C_PHP_L 1.465915e-02
## C_PHP_R 6.652227e-02
## C_PH_R 2.413927e-02
## C_R_L 7.296683e-01
## C_R_R 6.648078e-01
## EMC_L 3.059857e-03
## EMC_R 2.424058e-02
## FAT_L 3.380303e-05
## FAT_R 1.280932e-05
## IFOF_L 2.376973e-03
## IFOF_R 1.966622e-02
## ILF_L 4.039340e-02
## ILF_R 2.395401e-01
## MdLF_L 1.779880e-02
## MdLF_R 2.126197e-01
## PAT_L 7.147095e-03
## PAT_R 7.713464e-03
## SLF1_L 4.057589e-02
## SLF1_R 5.057219e-01
## SLF2_L 3.214374e-03
## SLF2_R 3.787585e-04
## SLF3_L 1.407435e-05
## SLF3_R 5.539809e-07
## UF_L 1.205368e-04
## UF_R 4.476156e-09
## VOF_L 5.287183e-01
## VOF_R 5.392905e-02print(fascicle_anova_fdr_association)## component p_FDR_corr
## 1 AF_L 0.003
## 2 AF_R 0.019
## 3 C_PHP_L 0.033
## 4 C_PH_R 0.043
## 5 EMC_L 0.01
## 6 EMC_R 0.043
## 7 FAT_L 0
## 8 FAT_R 0
## 9 IFOF_L 0.009
## 10 IFOF_R 0.039
## 11 MdLF_L 0.038
## 12 PAT_L 0.019
## 13 PAT_R 0.019
## 14 SLF2_L 0.01
## 15 SLF2_R 0.002
## 16 SLF3_L 0
## 17 SLF3_R 0
## 18 UF_L 0.001
## 19 UF_R 0#visreg
#sapply(fascicle_age_lm,visreg)########### age effect whole group
#isolate out only the depressed group and healthy group
dep_and_healthy_groups_for_ICD_analysis <- df_demo_and_fascicles[df_demo_and_fascicles$depGroupVar != 0,] #n depressed = 393, healthy = 713, total n = 1106
#### age histo ###
hist(dep_and_healthy_groups_for_ICD_analysis$PAT_AGE_AT_EXAM,
main = paste0("Age In Good Mimosa Group : n = ", dim(dep_and_healthy_groups_for_ICD_analysis)[1]), xlab = "Age", breaks = 10)
#lm
fascicle_age_lm <- lapply(fascicle_names, function(x)
{
lm(substitute(i ~ PAT_AGE_AT_EXAM, list(i = as.name(x))), data = dep_and_healthy_groups_for_ICD_analysis)
})
names(fascicle_age_lm) <- fascicle_names
#anova
fascicle_age_anova <- lapply(fascicle_age_lm, anova)
#fdr corrected
fascicle_age_anova_fdr <- fdr_anova_generic(fascicle_age_anova, 1)## p_anova
## AF_L 7.675516e-12
## AF_R 5.055895e-05
## C_FPH_L 8.258035e-03
## C_FPH_R 8.227520e-01
## C_FP_L 3.004545e-05
## C_FP_R 2.885625e-02
## C_PH_L 5.130225e-01
## C_PHP_L 2.952220e-04
## C_PHP_R 6.970269e-04
## C_PH_R 5.207577e-04
## C_R_L 4.424656e-03
## C_R_R 2.867211e-04
## EMC_L 2.219735e-06
## EMC_R 7.113004e-04
## FAT_L 2.001993e-14
## FAT_R 4.136943e-14
## IFOF_L 5.954939e-05
## IFOF_R 3.623102e-03
## ILF_L 1.245812e-03
## ILF_R 1.801121e-01
## MdLF_L 4.548818e-04
## MdLF_R 6.685245e-01
## PAT_L 1.850682e-10
## PAT_R 6.070195e-10
## SLF1_L 3.972055e-06
## SLF1_R 3.492555e-02
## SLF2_L 1.953684e-10
## SLF2_R 1.329061e-08
## SLF3_L 2.087589e-23
## SLF3_R 7.297727e-18
## UF_L 9.780502e-17
## UF_R 8.532133e-27
## VOF_L 4.644074e-02
## VOF_R 7.913203e-05
## CB_L 3.272553e-01
## CB_R 5.570468e-01
## ICP_L 3.210947e-01
## ICP_R 1.986896e-02
## MCP 9.800711e-01
## SCP 2.922375e-07
## V 5.601201e-01
## CNIII_L 7.783680e-01
## CNIII_R 2.838869e-01
## CNII_L 9.202604e-01
## CNII_R 2.219778e-02
## CNVIII_L 3.354313e-03
## CNVIII_R 3.001256e-02
## CNVII_L 4.433311e-02
## CNVII_R 5.494444e-02
## CNV_L 1.661940e-02
## CNV_R 4.313178e-01
## AR_L 8.703386e-02
## AR_R 1.019340e-01
## CBT_L 3.296664e-10
## CBT_R 1.979954e-11
## CPT_F_L 1.388156e-10
## CPT_F_R 3.354360e-07
## CPT_O_L 3.067032e-02
## CPT_O_R 2.004826e-02
## CPT_P_L 1.837730e-01
## CPT_P_R 6.035796e-01
## CS_A_L 1.286181e-23
## CS_A_R 1.171133e-16
## CS_P_L 1.278186e-01
## CS_P_R 2.064962e-03
## CS_S_L 2.418965e-09
## CS_S_R 8.237142e-10
## CST_L 3.107629e-02
## CST_R 5.050816e-02
## DRTT_L 2.976425e-06
## DRTT_R 1.333224e-07
## F_L 8.434985e-01
## F_R 5.960281e-01
## ML_L 1.114892e-01
## ML_R 6.331454e-01
## OR_L 4.213239e-03
## OR_R 4.525175e-03
## RST_L 7.897668e-17
## RST_R 2.603912e-14
## TR_A_L 8.847621e-24
## TR_A_R 1.985093e-17
## TR_P_L 1.335421e-01
## TR_P_R 1.495308e-02
## TR_S_L 7.172037e-08
## TR_S_R 1.213817e-06
## AC 1.543925e-03
## CC 8.497713e-04print(fascicle_age_anova_fdr)## component p_FDR_corr
## 1 AF_L 0
## 2 AF_R 0
## 3 C_FPH_L 0.014
## 4 C_FP_L 0
## 5 C_FP_R 0.045
## 6 C_PHP_L 0.001
## 7 C_PHP_R 0.002
## 8 C_PH_R 0.001
## 9 C_R_L 0.008
## 10 C_R_R 0.001
## 11 EMC_L 0
## 12 EMC_R 0.002
## 13 FAT_L 0
## 14 FAT_R 0
## 15 IFOF_L 0
## 16 IFOF_R 0.007
## 17 ILF_L 0.003
## 18 MdLF_L 0.001
## 19 PAT_L 0
## 20 PAT_R 0
## 21 SLF1_L 0
## 22 SLF2_L 0
## 23 SLF2_R 0
## 24 SLF3_L 0
## 25 SLF3_R 0
## 26 UF_L 0
## 27 UF_R 0
## 28 VOF_R 0
## 29 ICP_R 0.032
## 30 SCP 0
## 31 CNII_R 0.035
## 32 CNVIII_L 0.006
## 33 CNVIII_R 0.046
## 34 CNV_L 0.028
## 35 CBT_L 0
## 36 CBT_R 0
## 37 CPT_F_L 0
## 38 CPT_F_R 0
## 39 CPT_O_L 0.046
## 40 CPT_O_R 0.032
## 41 CS_A_L 0
## 42 CS_A_R 0
## 43 CS_P_R 0.004
## 44 CS_S_L 0
## 45 CS_S_R 0
## 46 CST_L 0.046
## 47 DRTT_L 0
## 48 DRTT_R 0
## 49 OR_L 0.008
## 50 OR_R 0.008
## 51 RST_L 0
## 52 RST_R 0
## 53 TR_A_L 0
## 54 TR_A_R 0
## 55 TR_P_R 0.026
## 56 TR_S_L 0
## 57 TR_S_R 0
## 58 AC 0.003
## 59 CC 0.002#fdr corrected only association cortex
fascicle_anova_w_fiber_mapping <- fascicle_age_anova[which(fascicle_bundle_mapping$name_vector == "association")]
fascicle_anova_fdr_association <- fdr_anova_generic(fascicle_anova_w_fiber_mapping,1)## p_anova
## AF_L 7.675516e-12
## AF_R 5.055895e-05
## C_FPH_L 8.258035e-03
## C_FPH_R 8.227520e-01
## C_FP_L 3.004545e-05
## C_FP_R 2.885625e-02
## C_PH_L 5.130225e-01
## C_PHP_L 2.952220e-04
## C_PHP_R 6.970269e-04
## C_PH_R 5.207577e-04
## C_R_L 4.424656e-03
## C_R_R 2.867211e-04
## EMC_L 2.219735e-06
## EMC_R 7.113004e-04
## FAT_L 2.001993e-14
## FAT_R 4.136943e-14
## IFOF_L 5.954939e-05
## IFOF_R 3.623102e-03
## ILF_L 1.245812e-03
## ILF_R 1.801121e-01
## MdLF_L 4.548818e-04
## MdLF_R 6.685245e-01
## PAT_L 1.850682e-10
## PAT_R 6.070195e-10
## SLF1_L 3.972055e-06
## SLF1_R 3.492555e-02
## SLF2_L 1.953684e-10
## SLF2_R 1.329061e-08
## SLF3_L 2.087589e-23
## SLF3_R 7.297727e-18
## UF_L 9.780502e-17
## UF_R 8.532133e-27
## VOF_L 4.644074e-02
## VOF_R 7.913203e-05print(fascicle_anova_fdr_association)## component p_FDR_corr
## 1 AF_L 0
## 2 AF_R 0
## 3 C_FPH_L 0.01
## 4 C_FP_L 0
## 5 C_FP_R 0.035
## 6 C_PHP_L 0.001
## 7 C_PHP_R 0.001
## 8 C_PH_R 0.001
## 9 C_R_L 0.006
## 10 C_R_R 0.001
## 11 EMC_L 0
## 12 EMC_R 0.001
## 13 FAT_L 0
## 14 FAT_R 0
## 15 IFOF_L 0
## 16 IFOF_R 0.005
## 17 ILF_L 0.002
## 18 MdLF_L 0.001
## 19 PAT_L 0
## 20 PAT_R 0
## 21 SLF1_L 0
## 22 SLF1_R 0.041
## 23 SLF2_L 0
## 24 SLF2_R 0
## 25 SLF3_L 0
## 26 SLF3_R 0
## 27 UF_L 0
## 28 UF_R 0
## 29 VOF_R 0#visreg
#sapply(fascicle_age_lm,visreg)########### sex effect whole group
#isolate out only the depressed group and healthy group
dep_and_healthy_groups_for_ICD_analysis <- df_demo_and_fascicles[df_demo_and_fascicles$depGroupVar != 0,] #n depressed = 393, healthhy = 713, total n = 1106
#### sex histo ###
hist(dep_and_healthy_groups_for_ICD_analysis$sex_binarized,
main = paste0("Sex In Good Mimosa Group : n (M/F)= ", sum(dep_and_healthy_groups_for_ICD_analysis$sex_binarized == 1), "/", sum(dep_and_healthy_groups_for_ICD_analysis$sex_binarized == 2)), xlab = "Sex", breaks = 2)
#lm
fascicle_sex_lm <- lapply(fascicle_names, function(x)
{
lm(substitute(i ~ osex, list(i = as.name(x))), data = dep_and_healthy_groups_for_ICD_analysis)
})
names(fascicle_sex_lm) <- fascicle_names
#anova
fascicle_sex_anova <- lapply(fascicle_sex_lm, anova)
#fdr corrected
fascicle_sex_anova_fdr <- fdr_anova_generic(fascicle_sex_anova, 1)## p_anova
## AF_L 9.013135e-01
## AF_R 6.042202e-01
## C_FPH_L 5.651633e-05
## C_FPH_R 8.947828e-01
## C_FP_L 8.940033e-01
## C_FP_R 2.547114e-01
## C_PH_L 2.229079e-01
## C_PHP_L 3.819814e-01
## C_PHP_R 4.042140e-03
## C_PH_R 8.421617e-04
## C_R_L 7.771252e-06
## C_R_R 3.815929e-03
## EMC_L 9.882825e-01
## EMC_R 3.279594e-01
## FAT_L 4.181022e-01
## FAT_R 3.361680e-01
## IFOF_L 2.913207e-01
## IFOF_R 1.492513e-01
## ILF_L 8.408734e-01
## ILF_R 4.847254e-02
## MdLF_L 8.735236e-01
## MdLF_R 8.581814e-01
## PAT_L 7.547800e-02
## PAT_R 3.639155e-02
## SLF1_L 1.251484e-02
## SLF1_R 4.032934e-02
## SLF2_L 9.072823e-01
## SLF2_R 5.662004e-02
## SLF3_L 8.436853e-01
## SLF3_R 3.052960e-01
## UF_L 3.691989e-01
## UF_R 2.233660e-01
## VOF_L 4.992865e-02
## VOF_R 9.332884e-01
## CB_L 6.972369e-03
## CB_R 6.829122e-04
## ICP_L 4.442021e-05
## ICP_R 6.964113e-07
## MCP 8.693158e-07
## SCP 9.896457e-04
## V 2.506036e-03
## CNIII_L 4.815513e-01
## CNIII_R 8.365707e-01
## CNII_L 1.682373e-05
## CNII_R 4.314508e-01
## CNVIII_L 1.638713e-04
## CNVIII_R 2.888902e-06
## CNVII_L 9.899615e-01
## CNVII_R 5.012845e-01
## CNV_L 8.446682e-01
## CNV_R 7.666538e-04
## AR_L 7.551024e-01
## AR_R 5.810518e-01
## CBT_L 5.504094e-01
## CBT_R 6.234223e-01
## CPT_F_L 6.771001e-01
## CPT_F_R 4.194010e-01
## CPT_O_L 3.024180e-02
## CPT_O_R 4.679213e-05
## CPT_P_L 1.407400e-01
## CPT_P_R 1.178473e-01
## CS_A_L 1.449502e-01
## CS_A_R 3.342726e-01
## CS_P_L 7.257309e-01
## CS_P_R 6.999654e-03
## CS_S_L 8.954306e-01
## CS_S_R 8.528411e-01
## CST_L 4.695392e-03
## CST_R 7.283609e-02
## DRTT_L 6.287781e-02
## DRTT_R 7.237371e-02
## F_L 9.099644e-02
## F_R 9.988956e-01
## ML_L 1.014903e-01
## ML_R 5.011675e-01
## OR_L 2.618914e-02
## OR_R 6.413325e-03
## RST_L 4.744324e-01
## RST_R 4.429868e-01
## TR_A_L 1.430677e-01
## TR_A_R 3.044128e-01
## TR_P_L 1.775434e-01
## TR_P_R 1.249664e-01
## TR_S_L 7.108138e-01
## TR_S_R 9.233697e-01
## AC 3.064604e-01
## CC 5.831278e-01print(fascicle_sex_anova_fdr)## component p_FDR_corr
## 1 C_FPH_L 0.001
## 2 C_PHP_R 0.022
## 3 C_PH_R 0.006
## 4 C_R_L 0
## 5 C_R_R 0.022
## 6 CB_L 0.03
## 7 CB_R 0.006
## 8 ICP_L 0.001
## 9 ICP_R 0
## 10 MCP 0
## 11 SCP 0.007
## 12 V 0.016
## 13 CNII_L 0
## 14 CNVIII_L 0.002
## 15 CNVIII_R 0
## 16 CNV_R 0.006
## 17 CPT_O_R 0.001
## 18 CS_P_R 0.03
## 19 CST_L 0.024
## 20 OR_R 0.03#fdr corrected only association cortex
fascicle_anova_w_fiber_mapping <- fascicle_sex_anova[which(fascicle_bundle_mapping$name_vector == "association")]
fascicle_anova_fdr_association <- fdr_anova_generic(fascicle_anova_w_fiber_mapping,1)## p_anova
## AF_L 9.013135e-01
## AF_R 6.042202e-01
## C_FPH_L 5.651633e-05
## C_FPH_R 8.947828e-01
## C_FP_L 8.940033e-01
## C_FP_R 2.547114e-01
## C_PH_L 2.229079e-01
## C_PHP_L 3.819814e-01
## C_PHP_R 4.042140e-03
## C_PH_R 8.421617e-04
## C_R_L 7.771252e-06
## C_R_R 3.815929e-03
## EMC_L 9.882825e-01
## EMC_R 3.279594e-01
## FAT_L 4.181022e-01
## FAT_R 3.361680e-01
## IFOF_L 2.913207e-01
## IFOF_R 1.492513e-01
## ILF_L 8.408734e-01
## ILF_R 4.847254e-02
## MdLF_L 8.735236e-01
## MdLF_R 8.581814e-01
## PAT_L 7.547800e-02
## PAT_R 3.639155e-02
## SLF1_L 1.251484e-02
## SLF1_R 4.032934e-02
## SLF2_L 9.072823e-01
## SLF2_R 5.662004e-02
## SLF3_L 8.436853e-01
## SLF3_R 3.052960e-01
## UF_L 3.691989e-01
## UF_R 2.233660e-01
## VOF_L 4.992865e-02
## VOF_R 9.332884e-01print(fascicle_anova_fdr_association)## component p_FDR_corr
## 1 C_FPH_L 0.001
## 2 C_PHP_R 0.027
## 3 C_PH_R 0.01
## 4 C_R_L 0
## 5 C_R_R 0.027#visreg
#sapply(fascicle_sex_lm,visreg)########### sex effect whole group
#isolate out only the depressed group and healthy group
dep_and_healthy_groups_for_ICD_analysis <- df_demo_and_fascicles[df_demo_and_fascicles$depGroupVar != 0,] %>%
group_by(EMPI) %>%
arrange(EXAM_DATE) %>%
slice(1) %>%
ungroup()
#### sex histo ###
hist(dep_and_healthy_groups_for_ICD_analysis$sex_binarized,
main = paste0("Sex In Good Mimosa Group : n (M/F)= ", sum(dep_and_healthy_groups_for_ICD_analysis$sex_binarized == 1), "/", sum(dep_and_healthy_groups_for_ICD_analysis$sex_binarized == 2)), xlab = "Sex", breaks = 2)
#lm
fascicle_sex_lm <- lapply(fascicle_names, function(x)
{
lm(substitute(i ~ osex, list(i = as.name(x))), data = dep_and_healthy_groups_for_ICD_analysis)
})
names(fascicle_sex_lm) <- fascicle_names
#anova
fascicle_sex_anova <- lapply(fascicle_sex_lm, anova)
#fdr corrected
fascicle_sex_anova_fdr <- fdr_anova_generic(fascicle_sex_anova, 1)## p_anova
## AF_L 0.76016866
## AF_R 0.70432464
## C_FPH_L 0.39710814
## C_FPH_R 0.50651311
## C_FP_L 0.75089245
## C_FP_R 0.38245705
## C_PH_L 0.76508017
## C_PHP_L 0.53572289
## C_PHP_R 0.21986115
## C_PH_R 0.16675373
## C_R_L 0.74684776
## C_R_R 0.98510035
## EMC_L 0.67591963
## EMC_R 0.95934604
## FAT_L 0.79676553
## FAT_R 0.78533447
## IFOF_L 0.52718875
## IFOF_R 0.49953762
## ILF_L 0.86346472
## ILF_R 0.36826195
## MdLF_L 0.86031175
## MdLF_R 0.95316243
## PAT_L 0.62927675
## PAT_R 0.28901821
## SLF1_L 0.21972016
## SLF1_R 0.92380902
## SLF2_L 0.91830059
## SLF2_R 0.63159106
## SLF3_L 0.75872385
## SLF3_R 0.51383433
## UF_L 0.26163586
## UF_R 0.95884074
## VOF_L 0.09899818
## VOF_R 0.81587309
## CB_L 0.66255878
## CB_R 0.46487164
## ICP_L 0.23480542
## ICP_R 0.03093436
## MCP 0.11753754
## SCP 0.05978804
## V 0.18209828
## CNIII_L 0.24040975
## CNIII_R 0.43976159
## CNII_L 0.07885099
## CNII_R 0.64998325
## CNVIII_L 0.01119782
## CNVIII_R 0.02602897
## CNVII_L 0.03663720
## CNVII_R 0.63267935
## CNV_L 0.93426333
## CNV_R 0.51554655
## AR_L 0.92307971
## AR_R 0.57759767
## CBT_L 0.98981327
## CBT_R 0.88717792
## CPT_F_L 0.84028707
## CPT_F_R 0.40796169
## CPT_O_L 0.51279080
## CPT_O_R 0.05224996
## CPT_P_L 0.40850920
## CPT_P_R 0.38229651
## CS_A_L 0.58061038
## CS_A_R 0.96713920
## CS_P_L 0.96221922
## CS_P_R 0.21589730
## CS_S_L 0.77000574
## CS_S_R 0.70882315
## CST_L 0.07888356
## CST_R 0.34518596
## DRTT_L 0.15361494
## DRTT_R 0.16953827
## F_L 0.19079818
## F_R 0.86860360
## ML_L 0.47879358
## ML_R 0.54279719
## OR_L 0.47143077
## OR_R 0.08866121
## RST_L 0.73363975
## RST_R 0.88102244
## TR_A_L 0.69298757
## TR_A_R 0.80245198
## TR_P_L 0.60525932
## TR_P_R 0.37224029
## TR_S_L 0.83941714
## TR_S_R 0.82316816
## AC 0.54605376
## CC 0.67415116print(fascicle_sex_anova_fdr)## [1] component p_FDR_corr
## <0 rows> (or 0-length row.names)#fdr corrected only association cortex
fascicle_anova_w_fiber_mapping <- fascicle_sex_anova[which(fascicle_bundle_mapping$name_vector == "association")]
fascicle_anova_fdr_association <- fdr_anova_generic(fascicle_anova_w_fiber_mapping,1)## p_anova
## AF_L 0.76016866
## AF_R 0.70432464
## C_FPH_L 0.39710814
## C_FPH_R 0.50651311
## C_FP_L 0.75089245
## C_FP_R 0.38245705
## C_PH_L 0.76508017
## C_PHP_L 0.53572289
## C_PHP_R 0.21986115
## C_PH_R 0.16675373
## C_R_L 0.74684776
## C_R_R 0.98510035
## EMC_L 0.67591963
## EMC_R 0.95934604
## FAT_L 0.79676553
## FAT_R 0.78533447
## IFOF_L 0.52718875
## IFOF_R 0.49953762
## ILF_L 0.86346472
## ILF_R 0.36826195
## MdLF_L 0.86031175
## MdLF_R 0.95316243
## PAT_L 0.62927675
## PAT_R 0.28901821
## SLF1_L 0.21972016
## SLF1_R 0.92380902
## SLF2_L 0.91830059
## SLF2_R 0.63159106
## SLF3_L 0.75872385
## SLF3_R 0.51383433
## UF_L 0.26163586
## UF_R 0.95884074
## VOF_L 0.09899818
## VOF_R 0.81587309print(fascicle_anova_fdr_association)## [1] component p_FDR_corr
## <0 rows> (or 0-length row.names)#visreg
#sapply(fascicle_sex_lm,visreg)########### sex effect whole group
#isolate out only the depressed group and healthy group
dep_and_healthy_groups_for_ICD_analysis <- df_demo_and_fascicles[df_demo_and_fascicles$depGroupVar != 0,] #n depressed = 393, healthhy = 713, total n = 1106
#### sex histo ###
hist(dep_and_healthy_groups_for_ICD_analysis$sex_binarized,
main = paste0("Sex In Good Mimosa Group : n (M/F)= ", sum(dep_and_healthy_groups_for_ICD_analysis$sex_binarized == 1), "/", sum(dep_and_healthy_groups_for_ICD_analysis$sex_binarized == 2)), xlab = "Sex", breaks = 2)
#lm
fascicle_sex_lm <- lapply(fascicle_names, function(x)
{
lmerTest::lmer(substitute(i ~ osex + (1|EMPI), list(i = as.name(x))), data = dep_and_healthy_groups_for_ICD_analysis)
})
names(fascicle_sex_lm) <- fascicle_names
#anova
fascicle_sex_anova <- lapply(fascicle_sex_lm, anova)
#fdr corrected
fascicle_sex_anova_fdr <- fdr_anova_generic(fascicle_sex_anova, 1)## p_anova
## AF_L 0.836344933
## AF_R 0.942953538
## C_FPH_L 0.273921238
## C_FPH_R 0.764606142
## C_FP_L 0.937894064
## C_FP_R 0.526951526
## C_PH_L 0.968717553
## C_PHP_L 0.860024375
## C_PHP_R 0.164260110
## C_PH_R 0.207906369
## C_R_L 0.231385825
## C_R_R 0.776243366
## EMC_L 0.655585844
## EMC_R 0.901641448
## FAT_L 0.951021127
## FAT_R 0.859412523
## IFOF_L 0.330173766
## IFOF_R 0.576241392
## ILF_L 0.725918489
## ILF_R 0.450731140
## MdLF_L 0.735707443
## MdLF_R 0.904182705
## PAT_L 0.803804103
## PAT_R 0.289044494
## SLF1_L 0.125134874
## SLF1_R 0.650836359
## SLF2_L 0.974339726
## SLF2_R 0.479266936
## SLF3_L 0.741490206
## SLF3_R 0.608541022
## UF_L 0.208365642
## UF_R 0.931226215
## VOF_L 0.178334192
## VOF_R 0.838137105
## CB_L 0.089880265
## CB_R 0.025005923
## ICP_L 0.021312746
## ICP_R 0.005610597
## MCP 0.031018380
## SCP 0.067100397
## V 0.035597496
## CNIII_L 0.456960041
## CNIII_R 0.671449807
## CNII_L 0.079534958
## CNII_R 0.584706577
## CNVIII_L 0.005761417
## CNVIII_R 0.001862260
## CNVII_L 0.813577735
## CNVII_R 0.633816459
## CNV_L 0.921605147
## CNV_R 0.123944970
## AR_L 0.946600030
## AR_R 0.699782657
## CBT_L 0.752623220
## CBT_R 0.987094959
## CPT_F_L 0.945794289
## CPT_F_R 0.433383627
## CPT_O_L 0.557873515
## CPT_O_R 0.053002959
## CPT_P_L 0.427083082
## CPT_P_R 0.350019541
## CS_A_L 0.352620858
## CS_A_R 0.749885123
## CS_P_L 0.970347000
## CS_P_R 0.207124601
## CS_S_L 0.943183658
## CS_S_R 0.697580473
## CST_L 0.099224241
## CST_R 0.268007469
## DRTT_L 0.254947295
## DRTT_R 0.201837466
## F_L 0.243818278
## F_R 0.495161362
## ML_L 0.471561436
## ML_R 0.527155451
## OR_L 0.448709261
## OR_R 0.157206215
## RST_L 0.626546176
## RST_R 0.709377283
## TR_A_L 0.408104676
## TR_A_R 0.646641597
## TR_P_L 0.642745344
## TR_P_R 0.464957031
## TR_S_L 0.977780042
## TR_S_R 0.848458068
## AC 0.823164406
## CC 0.865798352print(fascicle_sex_anova_fdr)## [1] component p_FDR_corr
## <0 rows> (or 0-length row.names)#fdr corrected only association cortex
fascicle_anova_w_fiber_mapping <- fascicle_sex_anova[which(fascicle_bundle_mapping$name_vector == "association")]
fascicle_anova_fdr_association <- fdr_anova_generic(fascicle_anova_w_fiber_mapping,1)## p_anova
## AF_L 0.8363449
## AF_R 0.9429535
## C_FPH_L 0.2739212
## C_FPH_R 0.7646061
## C_FP_L 0.9378941
## C_FP_R 0.5269515
## C_PH_L 0.9687176
## C_PHP_L 0.8600244
## C_PHP_R 0.1642601
## C_PH_R 0.2079064
## C_R_L 0.2313858
## C_R_R 0.7762434
## EMC_L 0.6555858
## EMC_R 0.9016414
## FAT_L 0.9510211
## FAT_R 0.8594125
## IFOF_L 0.3301738
## IFOF_R 0.5762414
## ILF_L 0.7259185
## ILF_R 0.4507311
## MdLF_L 0.7357074
## MdLF_R 0.9041827
## PAT_L 0.8038041
## PAT_R 0.2890445
## SLF1_L 0.1251349
## SLF1_R 0.6508364
## SLF2_L 0.9743397
## SLF2_R 0.4792669
## SLF3_L 0.7414902
## SLF3_R 0.6085410
## UF_L 0.2083656
## UF_R 0.9312262
## VOF_L 0.1783342
## VOF_R 0.8381371print(fascicle_anova_fdr_association)## [1] component p_FDR_corr
## <0 rows> (or 0-length row.names)#visreg
#sapply(fascicle_sex_lm,visreg)#sex by depression
#isolate out only the depressed group and healthy group
dep_and_healthy_groups_for_ICD_analysis <- df_demo_and_fascicles[df_demo_and_fascicles$depGroupVar != 0,] #n depressed = 393, healthhy = 713, total n = 1106
#lm
fascicle_sexdep_lm <- lapply(fascicle_names, function(x)
{
lm(substitute(i ~ osex*depGroupVar, list(i = as.name(x))), data = dep_and_healthy_groups_for_ICD_analysis)
})
names(fascicle_sexdep_lm) <- fascicle_names
#anova
fascicle_sexdep_anova <- lapply(fascicle_sexdep_lm, anova)
#fdr corrected
fascicle_sexdep_anova_fdr <- fdr_anova_generic(fascicle_sexdep_anova, 3)## p_anova
## AF_L 8.583610e-04
## AF_R 5.027177e-01
## C_FPH_L 4.735418e-01
## C_FPH_R 8.840278e-02
## C_FP_L 1.856520e-01
## C_FP_R 8.142388e-01
## C_PH_L 6.943242e-04
## C_PHP_L 1.743129e-02
## C_PHP_R 4.382374e-02
## C_PH_R 3.708596e-03
## C_R_L 6.008525e-03
## C_R_R 1.411951e-03
## EMC_L 6.031829e-02
## EMC_R 3.927887e-01
## FAT_L 4.456799e-02
## FAT_R 8.892097e-01
## IFOF_L 8.747430e-02
## IFOF_R 5.774558e-01
## ILF_L 1.117493e-02
## ILF_R 5.900012e-01
## MdLF_L 2.128987e-01
## MdLF_R 8.910367e-02
## PAT_L 2.263464e-01
## PAT_R 5.708667e-01
## SLF1_L 4.978482e-01
## SLF1_R 3.115217e-01
## SLF2_L 8.369327e-01
## SLF2_R 4.883136e-01
## SLF3_L 4.076455e-01
## SLF3_R 6.444182e-03
## UF_L 1.872461e-02
## UF_R 1.010824e-01
## VOF_L 6.317168e-01
## VOF_R 1.468084e-01
## CB_L 6.871696e-01
## CB_R 1.991249e-02
## ICP_L 2.188857e-04
## ICP_R 8.863064e-01
## MCP 3.426113e-01
## SCP 9.045840e-01
## V 1.905632e-02
## CNIII_L 3.390058e-01
## CNIII_R 4.529969e-01
## CNII_L 5.000222e-01
## CNII_R 5.553139e-01
## CNVIII_L 1.332072e-03
## CNVIII_R 3.205472e-01
## CNVII_L 9.585720e-01
## CNVII_R 6.223173e-01
## CNV_L 4.536984e-01
## CNV_R 1.492017e-01
## AR_L 1.541615e-01
## AR_R 1.603685e-01
## CBT_L 1.609530e-01
## CBT_R 7.329214e-01
## CPT_F_L 6.726400e-02
## CPT_F_R 7.428525e-01
## CPT_O_L 5.820521e-02
## CPT_O_R 7.743318e-01
## CPT_P_L 3.679313e-02
## CPT_P_R 4.398708e-01
## CS_A_L 3.878345e-04
## CS_A_R 9.168458e-03
## CS_P_L 6.775440e-02
## CS_P_R 3.254205e-01
## CS_S_L 3.737274e-02
## CS_S_R 7.909223e-01
## CST_L 3.693716e-02
## CST_R 2.878437e-01
## DRTT_L 5.501999e-03
## DRTT_R 3.943197e-01
## F_L 1.503966e-01
## F_R 2.310321e-01
## ML_L 4.223029e-03
## ML_R 7.498130e-01
## OR_L 2.192205e-02
## OR_R 7.293215e-01
## RST_L 1.942305e-03
## RST_R 2.423999e-02
## TR_A_L 2.833772e-05
## TR_A_R 1.274744e-03
## TR_P_L 4.488673e-02
## TR_P_R 3.961542e-01
## TR_S_L 4.085471e-02
## TR_S_R 8.633607e-01
## AC 2.616713e-01
## CC 1.930358e-01print(fascicle_sexdep_anova_fdr)## component p_FDR_corr
## 1 AF_L 0.015
## 2 C_PH_L 0.015
## 3 C_PH_R 0.032
## 4 C_R_L 0.04
## 5 C_R_R 0.015
## 6 SLF3_R 0.04
## 7 ICP_L 0.01
## 8 CNVIII_L 0.015
## 9 CS_A_L 0.011
## 10 DRTT_L 0.04
## 11 ML_L 0.033
## 12 RST_L 0.019
## 13 TR_A_L 0.002
## 14 TR_A_R 0.015#fdr corrected only association cortex
fascicle_anova_w_fiber_mapping <- fascicle_sexdep_anova[which(fascicle_bundle_mapping$name_vector == "association")]
fascicle_anova_fdr_association <- fdr_anova_generic(fascicle_anova_w_fiber_mapping,3)## p_anova
## AF_L 0.0008583610
## AF_R 0.5027176612
## C_FPH_L 0.4735418129
## C_FPH_R 0.0884027774
## C_FP_L 0.1856520164
## C_FP_R 0.8142387529
## C_PH_L 0.0006943242
## C_PHP_L 0.0174312871
## C_PHP_R 0.0438237362
## C_PH_R 0.0037085962
## C_R_L 0.0060085249
## C_R_R 0.0014119513
## EMC_L 0.0603182934
## EMC_R 0.3927887131
## FAT_L 0.0445679943
## FAT_R 0.8892096920
## IFOF_L 0.0874743001
## IFOF_R 0.5774558012
## ILF_L 0.0111749312
## ILF_R 0.5900011949
## MdLF_L 0.2128987477
## MdLF_R 0.0891036693
## PAT_L 0.2263464455
## PAT_R 0.5708667462
## SLF1_L 0.4978482291
## SLF1_R 0.3115217290
## SLF2_L 0.8369326751
## SLF2_R 0.4883136391
## SLF3_L 0.4076455231
## SLF3_R 0.0064441822
## UF_L 0.0187246103
## UF_R 0.1010823969
## VOF_L 0.6317168021
## VOF_R 0.1468084182print(fascicle_anova_fdr_association)## component p_FDR_corr
## 1 AF_L 0.015
## 2 C_PH_L 0.015
## 3 C_PH_R 0.032
## 4 C_R_L 0.037
## 5 C_R_R 0.016
## 6 SLF3_R 0.037#sex by depression
#isolate out only the depressed group and healthy group
dep_and_healthy_groups_for_ICD_analysis <- df_demo_and_fascicles[df_demo_and_fascicles$depGroupVar != 0,] #n depressed = 393, healthhy = 713, total n = 1106
#lm
fascicle_agedep_lm <- lapply(fascicle_names, function(x)
{
lm(substitute(i ~ PAT_AGE_AT_EXAM*depGroupVar, list(i = as.name(x))), data = dep_and_healthy_groups_for_ICD_analysis)
})
names(fascicle_agedep_lm) <- fascicle_names
#anova
fascicle_agedep_anova <- lapply(fascicle_agedep_lm, anova)
#fdr corrected
fascicle_agedep_anova_fdr <- fdr_anova_generic(fascicle_agedep_anova, 3)## p_anova
## AF_L 7.211620e-01
## AF_R 1.082882e-01
## C_FPH_L 3.077241e-01
## C_FPH_R 8.967246e-01
## C_FP_L 7.495799e-01
## C_FP_R 2.043195e-01
## C_PH_L 3.629991e-01
## C_PHP_L 6.003049e-03
## C_PHP_R 8.017692e-01
## C_PH_R 2.353131e-01
## C_R_L 2.330881e-01
## C_R_R 1.037015e-01
## EMC_L 6.033741e-01
## EMC_R 3.012157e-01
## FAT_L 8.602666e-01
## FAT_R 1.091803e-01
## IFOF_L 1.423418e-02
## IFOF_R 2.105918e-02
## ILF_L 1.929732e-02
## ILF_R 8.508492e-02
## MdLF_L 8.798349e-01
## MdLF_R 9.334369e-02
## PAT_L 1.411169e-01
## PAT_R 7.630269e-01
## SLF1_L 2.623452e-01
## SLF1_R 7.405326e-03
## SLF2_L 3.078077e-01
## SLF2_R 6.569123e-01
## SLF3_L 4.385438e-01
## SLF3_R 3.692092e-01
## UF_L 4.690172e-01
## UF_R 3.769388e-07
## VOF_L 3.457560e-02
## VOF_R 7.225215e-01
## CB_L 1.421889e-01
## CB_R 5.431225e-01
## ICP_L 7.072645e-01
## ICP_R 9.167468e-01
## MCP 7.146762e-01
## SCP 1.850616e-01
## V 8.332383e-01
## CNIII_L 1.373475e-01
## CNIII_R 5.603877e-02
## CNII_L 6.872842e-01
## CNII_R 8.632667e-02
## CNVIII_L 6.841645e-01
## CNVIII_R 4.099667e-01
## CNVII_L 1.688054e-01
## CNVII_R 1.846902e-01
## CNV_L 2.784274e-01
## CNV_R 5.123933e-01
## AR_L 5.933290e-01
## AR_R 9.748559e-01
## CBT_L 9.828514e-01
## CBT_R 2.990674e-01
## CPT_F_L 2.771966e-01
## CPT_F_R 2.534184e-01
## CPT_O_L 2.126409e-02
## CPT_O_R 7.323558e-02
## CPT_P_L 6.831947e-01
## CPT_P_R 7.514417e-02
## CS_A_L 7.220415e-01
## CS_A_R 5.290128e-01
## CS_P_L 2.684227e-01
## CS_P_R 2.997738e-01
## CS_S_L 5.940073e-01
## CS_S_R 4.223145e-01
## CST_L 3.060663e-01
## CST_R 4.370902e-01
## DRTT_L 6.668927e-01
## DRTT_R 9.381645e-02
## F_L 3.982578e-02
## F_R 4.465985e-01
## ML_L 2.230028e-01
## ML_R 6.738030e-03
## OR_L 2.968876e-03
## OR_R 1.184289e-02
## RST_L 4.827996e-01
## RST_R 2.787388e-01
## TR_A_L 7.359284e-01
## TR_A_R 4.978401e-01
## TR_P_L 3.061069e-01
## TR_P_R 1.352954e-01
## TR_S_L 8.370338e-01
## TR_S_R 3.133496e-01
## AC 1.753736e-01
## CC 5.076085e-02print(fascicle_agedep_anova_fdr)## component p_FDR_corr
## 1 UF_R 0#fdr corrected only association cortex
fascicle_anova_w_fiber_mapping <- fascicle_agedep_anova[which(fascicle_bundle_mapping$name_vector == "association")]
fascicle_anova_fdr_association <- fdr_anova_generic(fascicle_anova_w_fiber_mapping,3)## p_anova
## AF_L 7.211620e-01
## AF_R 1.082882e-01
## C_FPH_L 3.077241e-01
## C_FPH_R 8.967246e-01
## C_FP_L 7.495799e-01
## C_FP_R 2.043195e-01
## C_PH_L 3.629991e-01
## C_PHP_L 6.003049e-03
## C_PHP_R 8.017692e-01
## C_PH_R 2.353131e-01
## C_R_L 2.330881e-01
## C_R_R 1.037015e-01
## EMC_L 6.033741e-01
## EMC_R 3.012157e-01
## FAT_L 8.602666e-01
## FAT_R 1.091803e-01
## IFOF_L 1.423418e-02
## IFOF_R 2.105918e-02
## ILF_L 1.929732e-02
## ILF_R 8.508492e-02
## MdLF_L 8.798349e-01
## MdLF_R 9.334369e-02
## PAT_L 1.411169e-01
## PAT_R 7.630269e-01
## SLF1_L 2.623452e-01
## SLF1_R 7.405326e-03
## SLF2_L 3.078077e-01
## SLF2_R 6.569123e-01
## SLF3_L 4.385438e-01
## SLF3_R 3.692092e-01
## UF_L 4.690172e-01
## UF_R 3.769388e-07
## VOF_L 3.457560e-02
## VOF_R 7.225215e-01print(fascicle_anova_fdr_association)## component p_FDR_corr
## 1 UF_R 0#visreg
#sapply(fascicle_lm, visreg)
#sapply(fascicle_age_lm, visreg)
#sapply(fascicle_sex_lm, visreg)
#sapply(fascicle_sexdep_lm, visreg)
#sapply(fascicle_lm_unique, visreg)